ML080220282: Difference between revisions

From kanterella
Jump to navigation Jump to search
(Created page by program invented by StriderTol)
(StriderTol Bot change)
 
(One intermediate revision by the same user not shown)
Line 18: Line 18:


=Text=
=Text=
{{#Wiki_filter:Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Environmental Report 
{{#Wiki_filter:}}
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-iii License Renewal Application TABLE OF CONTENTS Section Page E.1 METHODOLOG Y..........................................................................................................
E-1 E.2 THREE MI L E ISLAND PRA MODEL ............................................................................
E-3 E.2.1 Level 1 TMI PRA Models .................................................................................
E-3 E.2.2 History of the TMI PRA Model s........................................................................
E-3 E.2.2.1 RISKMAN PRA Model s..................................................................
E-3 E.2.2.2 CAFTA PRA Model s.......................................................................
E-5 E.2.2.3 TMI Level 2 Mode l..........................................................................
E-7 E.2.2.3.1 Level 1 to Level 2 Interfac e........................................
E-8 E.2.2.3.2  Containment Event Tr e e Purpose ..........................
E-17 E.2.2.3.3 Release Categories and Source Term s...................
E-22 E.2.2.4 TMI Model Results .......................................................................
E-25 E.2.3 External Flooding Model
................................................................................
E-26 E.2.3.1 Core Dam a ge Sequence Identifica t io n.........................................
E-26 E.2.3.2 Level 2 Binning of External Flooding Scenario s...........................
E-27 E.2.3.2.1 Source Term Correlation f o r External Flood Sequences Over 305' ms l........................................
E-27 E.2.3.2.2 Source Term Correlation f o r External Flood Sequences Below 305' ms l......................................
E-28 E.2.3.2.3 External Flooding Binni n g Summary .......................
E-29 E.2.4 TMI-1 Peer Review S u mmar y........................................................................
E-29 E.3 LEVEL 3 PRA ANALYSI S...........................................................................................
E-31 E.3.1 Analysi s..........................................................................................................
E-31 E.3.2 Populatio n......................................................................................................
E-31 E.3.3 Econom y........................................................................................................
E-32 E.3.4 Food and Agricultur e......................................................................................
E-33 E.3.5 Nuclide Releas e.............................................................................................
E-33 E.3.6 Evacuation
.....................................................................................................
E-34 E.3.7 Meteorology
...................................................................................................
E-34 E.3.8 MACCS2 Result s...........................................................................................
E-36 E.4 BASELINE RISK MONETIZATIO N.............................................................................
E-37 E.4.1 Off-Site Ex p osure Cost-Ris k..........................................................................
E-37 E.4.2 Off-Site Economic Cost-Risk..........................................................................
E-37 E.4.3 On-Site Ex p osure Cost-Ris k..........................................................................
E-38 E.4.4 On-Site Cl e anup and Decontamination Cost-Ris k.........................................
E-39 E.4.5 Replacement Power C o st-Ris k......................................................................
E-40 E.4.6 Max i mum Averted Cost-Risk..........................................................................
E-41 E.4.6.1 Internal Events Maxim u m Averted Cost-Ris k...............................
E-41 E.4.6.2 External Flooding Events Maxim u m Averted Co s t-Ris k...............
E-42 E.4.6.3 Non-Flooding External Events Max i mum Avert e d Cost-Ris k....... E-43 E.4.6.4 TMI-1 M a x i mum Avert e d Cost-Risk
.............................................
E-44 E.5 PHASE I SAMA ANALYSIS
........................................................................................
E-45 E.5.1 SAMA Identificatio n........................................................................................
E-45 E.5.1.1 Level 1 TMI-1 Importance List Revie w.........................................
E-46 E.5.1.2 Level 2 TMI-1 Importance List Revie w.........................................
E-47 Environmental Report Appendix E  SAMA ANALYSIS Page E-iv Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE OF CONTENTS (CONTINUED)
Section Page E.5.1.3 Industry SAMA Ana l ysis Review
..................................................
E-47 E.5.1.3.1 Turkey Poin t.............................................................
E-48 E.5.1.3.2 ANO-1......................................................................
E-49 E.5.1.3.3 Palisades .................................................................
E-49 E.5.1.3.4 D.C. Cook ................................................................
E-50 E.5.1.3.5 Susquehann a...........................................................
E-51 E.5.1.3.6 Fitzpatrick
................................................................
E-52 E.5.1.3.7 Industry SAMA identification Summary
...................
E-52 E.5.1.4 TMI-1 IPE .....................................................................................
E-53 E.5.1.5 TMI-1 IPEEE
................................................................................
E-55 E.5.1.6 Use of External Events in the TMI-1 SAMA Identification proces s.........................................................................................
E-57 E.5.1.6.1 Internal Fir e s............................................................
E-58 E.5.1.6.2 Seismic Ev e nts ........................................................
E-64 E.5.1.6.3 High Wind Events
....................................................
E-74 E.5.1.6.4 External Floodin g.....................................................
E-76 E.5.1.6.5 Transportation and Nearby Facility Accidents
......... E-83 E.5.2 Phase I Screenin g..........................................................................................
E-85 E.6 PHASE II SAMA ANA L YSIS .......................................................................................
E-87 E.6.1 SAMA N u mber 1:  Enhance the SBO EDG with Auto Start and Load Capability
.......................................................................................................
E-88 E.6.1.1 Internal Events and Non-External Flooding Evaluatio n................
E-89 E.6.1.2 External Flooding Evaluatio n........................................................
E-91 E.6.1.3 Cost of Implementation
................................................................
E-93 E.6.1.4 Net Valu e......................................................................................
E-93 E.6.2 SAMA N u mber 2:  Install Damage Resistant Hi g h Temperature RCP Seals with a Portable 480V AC gen e rator for Ex t ended EFW operation
....... E-94 E.6.2.1 Internal Events and Non-External Flooding Evaluatio n................
E-94 E.6.2.2 External Flooding Evaluatio n........................................................
E-97 E.6.2.3 Cost of Implementation
................................................................
E-99 E.6.2.4 Net Valu e......................................................................................
E-99 E.6.3 SAMA N u mber 3:  Use NSCCW as an Alternate Cooling Source for the DHR Heat Exchangers (DH-C-1A/B) ...........................................................
E-100 E.6.3.1 Internal Events and Non-External Flooding Evaluatio n..............
E-100 E.6.3.2 External Flooding Evaluatio n......................................................
E-102 E.6.3.3 Cost of Implementation
..............................................................
E-103 E.6.3.4 Net Valu e....................................................................................
E-104 E.6.4 SAMA N u mber 4:  Provide alternate Power to HPI Pump Minimum Flow Recirculati o n Valves MU-V-36 and MU-V-3 7...............................................
E-104 E.6.5 SAMA N u mber 5:  Enhance Valves MU-V-7 6 A/B and MU-V-77A/B to Allow for Rapid Alignment Changes in Accident Condition s........................
E-106 E.6.5.1 Internal Events and Non-External Flooding Evaluatio n..............
E-107 E.6.5.2 External Flooding Evaluatio n......................................................
E-109 E.6.5.3 Cost of Implementation
..............................................................
E-110 E.6.5.4 Net Valu e....................................................................................
E-110 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-v License Renewal Application TABLE OF CONTENTS (CONTINUED)
Section Page E.6.6 SAMA N u mber 6:  Add Cross-Ties Within the Trains of Co o ling Systems - DHR, DHCCW, DHRW
..............................................................................
E-111 E.6.6.1 Internal Events and Non-External Flooding Evaluatio n..............
E-111 E.6.6.2 External Flooding Evaluatio n......................................................
E-114 E.6.6.3 Cost of Implementation
..............................................................
E-115 E.6.6.4 Net Valu e....................................................................................
E-116 E.6.7 SAMA N u mber 7:  Use Fire Service Water as a n Alternate Cooling Source for the ICCW Heat Exchangers
.......................................................
E-116 E.6.7.1 Internal Events and Non-External Flooding Evaluatio n..............
E-116 E.6.7.2 External Flooding Evaluatio n......................................................
E-118 E.6.7.3 Cost of Implementation
..............................................................
E-119 E.6.7.4 Net Valu e....................................................................................
E-119 E.6.8 SAMA N u mber 8:  Automate Reactor Coolant Pump Trip on High Motor Bearing Cooling temperatur e.......................................................................
E-120 E.6.8.1 Internal Events and Non-External Flooding Evaluatio n..............
E-120 E.6.8.2 External Flooding Evaluatio n......................................................
E-122 E.6.8.3 Cost of Implementation
..............................................................
E-123 E.6.8.4 Net Valu e....................................................................................
E-124 E.6.9 SAMA N u mber 9:  Proceduralize L o cal ADV operatio n...............................
E-124 E.6.10 SAMA N u mber 10:  Automate BWST Refill
.................................................
E-125 E.6.10.1 Internal Events and Non-External Flooding Evaluatio n..............
E-125 E.6.10.2 External Flooding Evaluatio n......................................................
E-127 E.6.10.3 Cost of Implementation
..............................................................
E-128 E.6.10.4 Net Valu e....................................................................................
E-128 E.6.11 SAMA N u mber 11:  Enhance Extreme External Flooding Mitigation Equipment to Address SBO and Loss of Seal Cooling Scena r ios ...............
E-129 E.6.11.1 Internal Events and Non-External Flooding Evaluatio n..............
E-129 E.6.11.2 External Flooding Evaluatio n......................................................
E-132 E.6.11.3 Cost of Implementation
..............................................................
E-133 E.6.11.4 Net Valu e....................................................................................
E-134 E.6.12 SAMA N u mber 12:  Use The DHR System as an Alternate Suction Source for HPI
.............................................................................................
E-134 E.6.12.1 Internal Events and Non-External Flooding Evaluatio n..............
E-134 E.6.12.2 External Flooding Evaluatio n......................................................
E-137 E.6.12.3 Cost of Implementation
..............................................................
E-138 E.6.12.4 Net Valu e....................................................................................
E-138 E.6.13 SAMA N u mber 13:  Change IA System Logic to Automatically Start IA-P-1A/B After a Low Voltage Trip in Conjunction wi t h an ESA S.......................
E-139 E.6.13.1 Internal Events and Non-External Flooding Evaluatio n..............
E-139 E.6.13.2 External Flooding Evaluatio n......................................................
E-141 E.6.13.3 Cost of Implementation
..............................................................
E-142 E.6.13.4 Net Valu e....................................................................................
E-142 E.6.14 SAMA N u mber 14:  Replace HPI Pump Cooling Alignment Valves with MOV s...........................................................................................................
E-143 E.6.14.1 Internal Events and Non-External Flooding Evaluatio n..............
E-143 E.6.14.2 External Flooding Evaluatio n......................................................
E-146 E.6.14.3 Cost of Implementation
..............................................................
E-147 E.6.14.4 Net Valu e....................................................................................
E-148 Environmental Report Appendix E  SAMA ANALYSIS Page E-vi Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE OF CONTENTS (CONTINUED)
Section Page E.6.15 SAMA N u mber 15:  Automatic Sw a p to Recirculation Mode
.......................
E-148 E.6.15.1 Internal Events and Non-External Flooding Evaluatio n..............
E-148 E.6.15.2 External Flooding Evaluatio n......................................................
E-150 E.6.15.3 Cost of Implementation
..............................................................
E-152 E.6.15.4 Net Valu e....................................................................................
E-153 E.6.16 SAMA N u mber 16:  Automate HPI I njection on Low Pressurizer Leve l....... E-153 E.6.16.1 Internal Events and Non-External Flooding Evaluatio n..............
E-153 E.6.16.2 External Flooding Evaluatio n......................................................
E-156 E.6.16.3 Cost of Implementation
..............................................................
E-157 E.6.16.4 Net Valu e....................................................................................
E-157 E.6.17 SAMA N u mber 17:  Auto Isolate St e am Genera t ors on High Steam Line Flow .............................................................................................................
E-158 E.6.17.1 Internal Events and Non-External Flooding Evaluatio n..............
E-158 E.6.17.2 External Flooding Evaluatio n......................................................
E-160 E.6.17.3 Cost of Implementation
..............................................................
E-160 E.6.17.4 Net Valu e....................................................................................
E-161 E.6.18 SAMA N u mber 18:  Provide the Capability to A l ign the Sta n dby Battery Charger and the 1A/1B DC Cross-tie from the MCR
...................................
E-161 E.6.18.1 Internal Events and Non-External Flooding Evaluatio n..............
E-161 E.6.18.2 External Flooding Evaluatio n......................................................
E-164 E.6.18.3 Cost of Implementation
..............................................................
E-166 E.6.18.4 Net Valu e....................................................................................
E-166 E.6.19 SAMA N u mber 19:  Ins t all Battery Backed Hydrogen Ignito r s or a Passive Hy drogen igniti o n syste m...............................................................
E-167 E.6.19.1 Internal Events and Non-External Flooding Evaluatio n..............
E-167 E.6.19.2 External Flooding Evaluatio n......................................................
E-169 E.6.19.3 Cost of Implementation
..............................................................
E-171 E.6.19.4 Net Valu e....................................................................................
E-171 E.6.20 SAMA N u mber 20:  Extend the High Pressure Boundary T hrough DHR Valve DH-V-3 for ISLOCA Isolation
.............................................................
E-172 E.6.20.1 Internal Events and Non-External Flooding Evaluatio n..............
E-172 E.6.20.2 External Flooding Evaluatio n......................................................
E-174 E.6.20.3 Cost of Implementation
..............................................................
E-175 E.6.20.4 Net Valu e....................................................................................
E-175 E.6.21 SAMA N u mber 21:  Ins t all Concrete Shields to Block Direct Pathways from the RPV to the Containment Wall and/or Direct Contai n m ent Flooding E a rly in Exter n al Flooding Scenarios ............................................
E-175 E.6.21.1 Internal Events and Non-External Flooding Evaluatio n..............
E-176 E.6.21.2 External Flooding Evaluatio n......................................................
E-178 E.6.21.3 Cost of Implementation
..............................................................
E-181 E.6.21.4 Net Valu e....................................................................................
E-181 E.6.22 SAMA N u mber 22:  Ins t all an inde p ende nt EFW s yste m............................
E-182 E.6.22.1 Internal Events and Non-External Flooding Evaluatio n..............
E-182 E.6.22.2 External Flooding Evaluatio n......................................................
E-185 E.6.22.3 Cost of Implementation
..............................................................
E-186 E.6.22.4 Net Valu e....................................................................................
E-186 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-vii License Renewal Application TABLE OF CONTENTS (CONTINUED)
Section Page E.6.23 SAMA N u mber 23:  Develop Alarm Response Procedures to Direct Operation of RR-V-5 on Low RBEC Flo w....................................................
E-187 E.6.23.1 Internal Events and Non-External Flooding Evaluatio n..............
E-187 E.6.23.2 External Flooding Evaluatio n......................................................
E-189 E.6.23.3 Cost of Implementation
..............................................................
E-190 E.6.23.4 Net Valu e....................................................................................
E-190 E.6.24 SAMA N u mber 24:  Ins t all Damage Resistant High Temperature RCP Seals with a Diesel Engine as an al t ernate Drive for an EFW Pump and a Portable 480V AC ge n erator for Extended EFW operatio n.......................
E-191 E.6.24.1 Internal Events and Non-External Flooding Evaluatio n..............
E-191 E.6.24.2 External Flooding Evaluatio n......................................................
E-193 E.6.24.3 Cost of Implementation
..............................................................
E-194 E.6.24.4 Net Valu e....................................................................................
E-194 E.6.25 SAMA number 25:  Ins t all an Additional ED G..............................................
E-195 E.6.25.2 Non-Exter n al Flooding Evaluatio n..............................................
E-195 E.6.25.3 External Flooding Evaluatio n......................................................
E-196 E.6.25.3 Cost of Implementation
..............................................................
E-197 E.6.25.4 Net Valu e....................................................................................
E-197 E.6.26 SAMA number 26:  Reroute Cables so that they do not pass Over Ignition So urces in Fire Zone CB-FA-2E (West Inverter Ro o m) or Wrap them in Fire Proof Materia l...........................................................................
E-198 E.6.27 SAMA number 27:  Improve the 480V aC Load Center Weld s....................
E-202 E.6.27.1 Cost of Implementation
..............................................................
E-206 E.6.27.2 Net Valu e....................................................................................
E-207 E.6.28 SAMA number 28:  Improve the D e cay Heat S e rvice Cooler (DC-C-2A
/B) Anchorage s..................................................................................................
E-207 E.6.28.1 Cost of Implementation
..............................................................
E-210 E.6.28.2 Net Valu e....................................................................................
E-210 E.6.29 SAMA number 29:  Replace EDG Ground Resistor s...................................
E-211 E.6.29.1 Cost of Implementation
..............................................................
E-213 E.6.29.2 Net Valu e....................................................................................
E-213 E.6.30 SAMA number 30:  Improve Diesel Fire Pump Fuel Oil Tank and Battery Rack Supports
.............................................................................................
E-214 E.6.30.1 Cost of Implementation
..............................................................
E-215 E.6.30.2 Net Valu e....................................................................................
E-215 E.6.31 SAMA number 31:  Modify Specific Containment Penetration MOVs to FAIL Closed
.................................................................................................
E-215 E.6.32 SAMA N u mber 32:  Pre-stage severe external flooding equipmen t.............
E-216 E.6.32.1 Internal Events and Non-External Flooding Evaluatio n..............
E-218 E.6.32.2 External Flooding Evaluatio n......................................................
E-219 E.6.32.3 Cost of Implementation
..............................................................
E-222 E.6.32.4 Net Valu e....................................................................................
E-222 E.6.33 SAMA N u mber 33:  Inc r ease the Fl o od Protecti o n height ...........................
E-222 E.6.33.1 Internal Events and Non-External Flooding Evaluatio n..............
E-224 E.6.33.2 External Flooding Evaluatio n......................................................
E-226 E.6.33.3 Cost of Implementation
..............................................................
E-227 E.6.33.4 Net Valu e....................................................................................
E-227 Environmental Report Appendix E  SAMA ANALYSIS Page E-viii Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE OF CONTENTS (CONTINUED)
Section Page E.7 UNC E RTA I NTY ANAL YSI S......................................................................................
E-228 E.7.1 95 th Percen t ile PRA Result s.........................................................................
E-228 E.7.1.1 Phase I Impact
...........................................................................
E-229 E.7.1.2 Phase II Impact
..........................................................................
E-229 E.7.2 BWST Refill capabilit y..................................................................................
E-231 E.7.3 MACCS2 I n put variation s.............................................................................
E-234 E.7.3.1  Meteorological Sensitivi t y...........................................................
E-235 E.7.3.2 Population Sensitivity
.................................................................
E-235 E.7.3.3 Evacuation Sensitivity
................................................................
E-236 E.7.3.4 Radioactive Release Sensitivity
.................................................
E-236 E.7.3.5 Intermediate Phase Duration Sensiti v ity ....................................
E-236 E.7.3.6 Impact on SAMA Ana l ysi s..........................................................
E-237 E.7.4 Extr e m e Flooding Mitigation
........................................................................
E-238 E.7.4.1 Phase I impac t............................................................................
E-238 E.7.4.2 Phase II impac t...........................................................................
E-239 E.7.5 Sensitivity Analysis: Impact of Implementing SAMA 3 2...............................
E-241 E.7.5.1 Analysis Proces s........................................................................
E-241 E.7.5.2 Results .......................................................................................
E-243 E.7.6 Sensitivity Analysis: Impact of SECPOP Error Corrections
.........................
E-244 E.7.6.1 Error #1 ......................................................................................
E-244 E.7.6.2 Error #2 ......................................................................................
E-244 E.7.6.3 Error #3 ......................................................................................
E-245 E.7.6.4 Impact on TMI MAC R.................................................................
E-245 E.7.6.5 Impact on Individual SAMA Calculation s....................................
E-247 E.8 CONCLUS I ON S........................................................................................................
E-251 E.9 FIGURES ..................................................................................................................
E-253 E.10 TABLES ....................................................................................................................
E-256 E.11 REFERENCE S..........................................................................................................
E-362 ADD E NDUM 1 TO AT TACHMENT E SELEC T E D PR EVIOUS INDUSTRY SAMA S......................................................................................................................
E-3 6 9 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-ix License Renewal Application TABLE OF CONTENTS (CONTINUED)
List of Tables Table Page Table E.2-1 Three Mile Island PRA Model Summary
...........................................................
E-256 Ta b l e E.2-2 Core Melt B i ns ...................................................................................................
E-257 Table E.2-3 Core Melt B i n Assignments for Level 1  Transient Core Dam a ge State s..........
E-257 Table E.2-4 Core Melt B i n Assignments for Level 1  LOOP-SBO Core D a mage Stat e s...... E-258 Table E.2-5 Core Melt B i n Assignments for Level 1 Very Sm a ll  LOCA C o re Damage States ............................................................................................................
E-259 Table E.2-6 Core Melt B i n Assignments for Level 1  Small LOCA Core D a mage Stat e s..... E-260 Table E.2-7 Core Melt B i n Assignments for Level 1 Medium L O CA Core Damage St a te s.. E-260 Table E.2-8 Core Melt B i n Assignments for Level  1 Large LOCA Core D a mage Stat e s..... E-260 Table E.2-9 Core Melt B i n Assignments for Level 1  SGTR Core Damage State s...............
E-261 Table E.2-10 Core Melt Bin Assignments for Level 1 Steamline Breaks Upstream MSIVs Core Dam a ge State s.....................................................................................
E-261 Table E.2-11 Core Melt Bin Assignments for Level 1 Steamline Breaks Downstream MSIVs Core Damage S t ates .........................................................................
E-262 Table E.2-12 Core Melt Bin Assignments for Level 1 ATWS  Core Dam a ge States ............
E-262 Table E.2-13 Core Melt Bin Assignments for Level 1 ISLOCA Core Dam a ge States...........
E-263 Ta b l e E.2-1 4 Containment Safeguards/Isola t ion State .........................................................
E-263 Table E.2-15 Containment Event Tree Top Ev e nts ..............................................................
E-264 Table E.2-16 Individual Release Category Definition s..........................................................
E-265 Table E.2-17 Summary of Representative MAAP Sequences for TMI-1 S o urce Terms.......
E-268 Table E.2-18 TMI-1 So urce Term Summar y.........................................................................
E-269 Table E.2-19 TMI-1 Initiating Event Contributions to CD F....................................................
E-273 Table E.2-20 TMI-1 Top Initiating Event Contributions for Each Release Categor y.............
E-273 Table E.2-21 External Flooding CDF Summar y....................................................................
E-274 Table E.2-22 CET Node Binning Characteristic s..................................................................
E-275 Table E.2-23 Flood Sequence Source Term Frequencie s....................................................
E-276 Table E.2-24 TMI Peer Review S u mmary Overall Assessment
...........................................
E-277 Table E.3-1 Estimated Population Distribution Within a 50-Mile Radius of Three Mile Island, Year 2034 ..........................................................................................
E-278 Table E.3-2 Estimated Population Distribution Within a 50-Mile Radius of Three Mile Island, Year 2034 ..........................................................................................
E-279 Table E.3-3 Three Mile Island MACCS2 Core I n ventory ......................................................
E-280 Table E.3-4 MACCS2 Release Cat e gories vs.
T hree Mile Island  Release Categories
....... E-281 Table E.3-5 MACCS2 Base Case Mean Result s..................................................................
E-281 Table E.3-6 Release Category Specific MACCS2 Base Case Mean Results
......................
E-282 Table E.3-7 External Flooding Base Case Mean Result s.....................................................
E-282 Table E.5-1 Level 1 Im p ortance List Review ........................................................................
E-283 Table E.5-2 Level 2 Im p ortance List Review ........................................................................
E-307 Table E.5-3 Phase I SAM A...................................................................................................
E-3 2 5 Table E.5-4 Phase II SAM A..................................................................................................
E-3 4 5 Table E.8-1 Summay of Cost Beneficial SAMAs
..................................................................
E-356 Environmental Report Appendix E  SAMA ANALYSIS Page E-x Three Mile Island Nuclear Station Unit 1 License Renewal Application Acronyms Used in Attachment E AC alternating current ADV atmospheric dump valve AFW auxiliary feedwater ATWS anticipated transient without scram BWR boiling water reactor BWST borated water storage tank CCF common cause failure CDF core damage frequency CET containment event tree CFS cavity flooding system CR control room CRD control rod drive CS containment spray CST condensate storage tank DA data analysis DC direct current DCH direct containment heating DG diesel generator DHCCW decay heat closed cooling water DPD dollar per dollar ECCS emergency core cooling system EDG emergency diesel generator EFW emergency feedwater EOP emergency operating procedures EPRI electric power research institute EPZ emergency planning zone FIVE fire induced vulnerability evaluation FP fire protection FPS fire protection system F-V Fussell-Vesely GIS geographic information system gpm gallons per minute HEP human error probability HPCI high pressure coolant injection HPI high pressure injection HPME high pressure melt ejection HPSI high pressure safety injection HRA human reliability analysis HVAC heating ventilation and  air-conditioning system IA Instrument air ICS instrumentation and control system IE initiating event IPE individual plant examination IPEEE individual plant examination - external events ISLOCA interfacing system LOCA JHEP joint human error probability LERF large early release frequency LLNL Lawrence Livermore National Labs LOCA loss-of-coolant accident Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-xi License Renewal Application Acronyms Used in Attachment E LOOP loss of off-site power LPR low pressure recirc LPSI low pressure safety injection MAAP modular accident analysis program MACCS2 melcor accident consequences code system, version 2 MACR maximum averted cost-risk MCC motor control center MCR main control room MET meteorological MSIV main steam isolation valve msl mean sea level MTC moderator-temperature co-efficient NPSH net positive suction head NRC U.S. nuclear regulatory commission NSCCW Nuclear Services Closed Cooling Water NSRW Nuclear Service River Water OECR off-site economic cost risk OSP off-site power OTSG once through steam generator PDS plant damage state PORV pressure operated relief valve PRA probabilistic risk analysis PSA probabilistic safety assessment PTS pressurized thermal shock PWR pressurized water reactor RBNC reactor building normal cooling RCIC reactor core isolation cooling RCP reactor coolant pump RCS reactor coolant system RDR real discount rate RHR residual heat removal RHRSW residual heat removal service water RPV reactor pressure vessel RRW risk reduction worth RSP remote shutdown panel SAMA severe accident mitigation alternative SBO station blackout SDP significance determination process SGTR steam generator tube rupture SPRA seismic PRA SRV safety relief valve SSC systems structures & components SSES Susquehanna Steam Electric Station SSHR secondary side heat removal SSRW secondary service river water ST structural response SW service water TD EFW turbine driven EFW TMI Three Mile Island 
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-1 License Renewal Application Appendix E Severe Accident Mitigation Alternatives The severe accident mitigation alter n atives (SAMA) analysis discussed in Section 4.
2 0 of the Environmental Report is presented below.
E.1 METHODOLOGY The methodology selected for this analysis involves identifying SAMA candidates that have the highest potential for reducing plant risk below the currently acceptably-low levels 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 off-site economic cost-risk. These values provide a measure of both the likelihood and consequences of a core damage event. The SAMA process consists of the following steps:
* TMI-1 Probabilistic Risk Assessment (PRA) Model - Use the TMI-1 Internal Events PRA model as the basis for t he analysis (Section E.2
). Incorpor a te external events contributions as descri be d in Sectio ns E.4.6 and E.6.
* Level 3 PRA Analysis - Use TMI-1 Level 1 (CDF) and Level 2 (Containment Response) Internal Events PRA output and site-specific meteorology, demographic, land use, and emergency response data as input in performing a Level 3 (offsite consequences) PRA using the MELCOR Accident Cons e quences C o de System Version 2 (M ACCS2) (Section E.3).
* Baseline Risk Monetization - Use U.S. Nuclear Regulatory Commission (NRC) regulatory analysis techniques to calculate the monetary value of taking no further action to reduce the consequences of potential severe accidents for TMI-1. This becomes the maximum averted
 
cost-risk (MACR) that is possible (Section E.4
).
* Phase I SAMA Analysis - Identify potential SAMA candidates based on the TMI-1 PRA, Individual Plant Examination - External Events (IPEEE), and documentation from the industry and NRC. Screen out Phase I SAMA candidates that are not applicable to the TMI-1 design or are of low benefit in pressurized water reactors (PWRs) such as TMI-1, Environmental Report Appendix E  SAMA ANALYSIS Page E-2 Three Mile Island Nuclear Station Unit 1 License Renewal Application candidates that have already been implemented at TMI-1 or whose benefits have been achieved at TMI-1 using other means, and candidates whose estimated cost exceeds the possible MACR (Section E.5
).
* Phase II SAMA Analysis - Calculate the risk reduction attributable to each remaining SAMA candidate and compare to a more detailed cost analysis to identify the net cost-benefit.
PRA insights are also u s ed to scre e n SAMA ca n didates in t h is phase (S e c tion E.6).
* Uncertainty Analysis - Evaluate how changes in the SAMA analysis assumptions might affect the c o st-benefit evaluation (S e c tion E.7).
* Conclusions - Summar i ze results a nd identify conclusio n s (Section E.8
). The steps outlined above are described in more detail in the subsections of this appendix. The graphic below summarizes the high-level steps of the SAMA process. Initial SAMA ListApplicable to Plant?YesScreened No NoScreened YesDoes the SAMA affect a risk significant system?YesScreened NoImplementation cost greater than cost-risk reduction?
NoScreenedYesRetain for potential implementationIs Implementation cost greater than screening cost?Phase IAnalysisPhase IIAnalysis Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-3 License Renewal Application E.2 THREE MILE ISLAND PRA MODEL This section provides a summary of the Three Mile Island (TMI) PRA model used to support the SAMA analysis and the changes that have been made to the model since the individual plant examination (IPE). The external events models are not specifically discussed in this section. E.2.1 LEVEL 1 TMI PRA MODELS The TMI 2004 Revision 2 Level 1 PRA model (Exelon 2007a), the most recent model, calculated a Core Damage Frequency (CDF) of 2.37E-5/yr and a value for Large Early Release Frequency (LERF) of 3.02E-06/yr.
Table E.2-1 summarizes the historical values for previous TMI models and their calculated values for CDF and LERF. E.2.2 HISTORY OF THE TMI PRA MODELS E.2.2.1 RISKMAN PRA MODELS The TMI-1 Level I PRA was updated in late 1989 and 1990 to revise the internal events portion of the Level I PRA that was initially completed in 1987. The updates were undertaken to reflect changes in plant design and procedures made sinc e 1987 and to fulfill the requirements of NRC Generic Letter 88-20, "Individual Plant Examinations". In conformance with those requirements the major objectives of the PRA update were to: 1. Further develop an appreciation of severe accident behavior. 2. Build on the understanding of the most likely severe accident sequences that could occur at TMI-1. 3. Improve the quantitative understanding of the overall probabilities of core damage. The updates were conducted in a manner that maximized the use of in-house personnel. Plant and Support PRA analysts and engineers and operators who were familiar with the details of the design, controls, procedures, and system configurations were directly involved in the analysis as well as the technical review. Various consultants have assisted the TMI-1 PRA staff in the update by providing expertise in the plant model revisions and in various special analyses. An additional objective of the study was to build on existing in-house PRA expertise and to develop tools for ongoing risk management activities after the completion of the PRA update.
Environmental Report Appendix E  SAMA ANALYSIS Page E-4 Three Mile Island Nuclear Station Unit 1 License Renewal Application The IPE submittal (December 1992 model) was based on the plant as it was configured in 1991. The RISKMAN models of 2000, 2001, and 2003 were based on the plant as it was configured in 1998. The 2001 model, which was known as L2RV2, was the one primarily used for configuration risk management purposes. Al though the 2003 RISKMAN model (ABSA) (ABS 2003) was not officially used for configuration risk management purposes, it provided the basis for later PRA models that were converted to CAFTA. The list below shows the major plant and procedure changes made since 1987 that were significant to these RISKMAN PRA models. Most of these changes were made as a result of insights gained from the original 1987 PRA
 
model. 1. Addition of an alternate AC source (TMI Unit 2 diesel generator) with the ability to tie into either division of 1E power. 2. Installation of improved reactor coolant pump seals that reduce the likelihood of seal failure under loss of injection and cooling conditions. 3. Modification of the power supplies to the ICS that eliminates loss of 120V AC bus ATA power as an initiating event. 4. Addition of an air compressor, air dryer and filters that improves the reliability of the instrument air system. 5. Change to procedure for loss of air that directs the operator to manually open RCP seal return valve (MU-V-20). This assures continuation of RCP seal injection during loss of air scenarios. 6. Modification of the power supplies to the "B" HPI pump and its associated lube oil pumps that assures they both are supplied from the same source of power. This reduces the chances of pump failure if power is lost. 7. Relocation of the control switches for HPI pump min-recirc valves (MU-V-36 and 37) from the back panels to the control room console. This reduces the likelihood of operator failure to re-establish min-recirc after throttling HPI and thus reduces the likelihood of pump damage and consequent loss of RCP seal injection. 8. Changes to procedures for loss of river water events that direct the operator to alternate make-up pumps to utilize the heat capacity of the DHCC system as a heat sink for pump cooling, and if necessary to cross-connect firewater to the DHCC heat exchangers. This reduces the likelihood that a loss of river water intake event would lead to loss of RCP seal injection. 9. A change to torque switch settings for DHR isolation valves (DH-V-4A & B) that improves the ability of the valves to be closed against a high differential pressure. This reduces the likelihood of an interfacing system LOCA (ISLOCA) through these valves. 10. Addition of a diverse scram system to reduce the likelihood of an ATWS.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-5 License Renewal Application
: 11. Modification of the balance of plant power supply distribution to minimize the chances of trip due to loss of DC train A. 12. Replacement of the analog turbine control system with a digital control system. For a closeout summary of the recommendations of the 1987 PRA, which includes most of these changes, see GPUN letter to NRC of February 22, 1990 (H.D. Hukill to NRC, #C311 2012). Two independent reviews of the December 1992 update were conducted:  one by an independent in-house group consisting of managers of key organizations, and one by an external consultant. The purpose of the independent in-house review was to ensure the accuracy of the documentation and to validate the PRA process and its results. The external consultant review was conducted to ensure that proper PRA techniques were employed and that key iss u es were addressed. The results of t h ese reviews were provi d ed in Appendix D of Reference (GPU 1992). E.2.2.2 CAFTA PRA MODELS As mentioned above, the ABSA 2003 RISKMAN model provided the basis for a conversion to a Level 1 CAFTA software model in 2004. The ABSA model addressed significant findings from the TMI PRA Peer Certification ("A" and "B" F&Os). The CAFTA conversion improved the details in several system models, accident sequence event trees, and updated the initiating event and component failure/unavailability rates. Changes made to the PRA were done to
 
support procedural requirements for a periodic update to support risk informed applications and configuration risk management. The 2004 Revision 0 model was never offi cially implemented, with the 2004 Revision 1 model being the official model of record since June 2005 (Exelon 2005b). The 2004 Revision 1 upgrade was performed to correct errors discovered subsequent to the conversion to CAFTA (the 2004 Rev. 0 model) and enhance the model for use in configuration risk management. Key changes made to the TMI PRA since the RISKMAN TMIL2RV2 model of 2001 are listed below. Changes made to the model for the interim 2003 update (TMIABSA) are so designated: 1. [TMIABSA] Incorporated updated values for initiating event frequencies, component failure rates, unavailability, and common cause factors. 2. [TMIABSA] Updated the Level 2 model assumptions to reflect progress in industry research and understanding from the last several years.
Environmental Report Appendix E  SAMA ANALYSIS Page E-6 Three Mile Island Nuclear Station Unit 1 License Renewal Application 3. [TMIABSA] Updated entire HRA using EPRI HRA Calculator. 4. [TMIABSA] Re-evaluated success criteria and operator action timing using results from updated thermal-hydraulic (MAAP) analyses. 5. [TMIABSA] Refined the screening analysis previously used for internal flooding 6. Converted the Model from a RISKMAN linked event tree model to a CAFTA single top event fault tree model. During this c onversion each event tree was modified. 7. Enhanced the following system models:  Main Feedwater and Main Steam as they relate to OTSG isolation for SGTR and secondary line breaks. 4KV/480V AC power was updated to include individual fault trees for 480V buses and MCCs. Updated common cause data to NUREG/CR-5497. Added logic to evaluate system availability following offsite power recovery. 8. Performed a detailed operator action dependency analysis. Developed Joint Human Error Probability (JHEP) basic events and added them to the PRA model as appropriate. 9. Performed numerous minor updates and enhancements to the model, which included changes to basic event names and probabilities, nodal logic for most event trees, and the logic for several top events and systems. These changes are all described in
 
Attachment C of Reference (Exelon 2005b). The 2004 Revision 2 model, upon which this SAMA analysis is based, superseded the Revision 1 model in 2007. Key changes and modifications included revision of common cause failure events and their probabilities using the data provided in (NRC 1998b). A summary listing of the changes and improvements made since the 2004 Revision 1 model is listed below: 1. New basic events were added to the PRA model for common cause failures of the batteries, inverters, battery chargers, pressurizer safety valves, and steam generator atmospheric dump valves. 2. New maintenance unavailability events were added to include maintenance on various components not previously modeled. Various old maintenance unavailability basic event names were replaced with new names to adopt a more consistent naming scheme. The time period for the maintenance unavailability data was the same as that used for the Revision 1 model (1998 to 2001). 3. Revision of fault tree logic for the makeup pumps in support of the high pressure injection and reactor coolant pump (RCP) seal injection functions.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-7 License Renewal Application
: 4. Uncertainty data was added to the TMI database files, which identified error factors and the distribution type (lognormal) for type code assignments and unique basic events, such as maintenance unavailabilities, common cause events, and human event probability (HEP) actions. 5. Addition of new HEPs for controlling emergency feedwater, cooldown of the reactor coolant system (RCS), and steam generator isolation. 6. New HEP dependencies were identified and JHEP events created to account for the addition of new HEPs within the PRA model for the electrical DC and Nuclear Service River Water (NSRW) systems. 7. The loss of offsite power (LOOP) initiating event frequency was revised to be 4.48E-2 per year based on a generic prior distribution with a Bayesian update using data from 1997 to 2003. 8. Since low pressure recirculation (LPR) was considered a viable option given the success of cooling down the RCS, the event tree for Very Small LOCAs was modified to include a low pressure recirculation node. 9. New logic was added to account for makeup pump lube oil pump run failures and power supply dependencies, since failure of both lube oil pumps will fail their respective
 
makeup pump. 10. New logic was added to the Decay Heat River Water system fault trees to account for the fact that the decay heat river pumps are running about 50% of the time (25% for each train), and thus would not need to start. 11. Improvements were made to the logic for the NSRW system that credits use of the Secondary Service River Water (SSRW) system to recover failures of NSRW, e.g.,
failure of the NSRW pumps. Also, adjustments were made to take credit for recovery of certain loss of NSRW initiators (%LNR); since it was found that a 73% contribution toward initiating event %LNR was recoverable by use of the NSRW-SSRW cross-tie. 12. Inverters 1E and 1F in the 120V AC vital electrical system were credited with the ability to provide a backup power supply for the normally in-service inverters.
Section E.2.4 summarizes the peer reviews performed on the TMI-1 PRA models. E.2.2.3 TMI LEVEL 2 MODEL The Level 2 model used for the SAMA analysis is linked to the core damage sequences from the CAFTA 2004 Revision 2 Level 1 model described above. The methodology for the Containment Event Tree (CET) solution, the CET quantification, and source term development were based on the TMI IPE Level 2 analysis of 1993, which was originally based on the Oconee PRA Level 2 analysis. Oconee and TMI-1 designs were compared to identify any significant differences in plant characteristics. Then, the Oconee CET model and its quantification were modified to reflect these differences, as well as develop a plant specific model for TMI-1. TMI-1 Environmental Report Appendix E  SAMA ANALYSIS Page E-8 Three Mile Island Nuclear Station Unit 1 License Renewal Application specific analyses using the MAAP code were performed to further enhance the Oconee model and verify its applicability to TMI-1. The TM I CAFTA Level 2 model of 2007 and CET used for this SAMA analysis are fully described in the TMI-PRA-001 (Exelon 2007b). E.2.2.3.1 Level 1 to Level 2 Interface In order to determine the consequences of a reactor accident, the sequences identified as leading to core damage must be analyzed in terms of various phenomena that can occur in-plant (i.e., inside the reactor vessel and containment). This involves carrying the sequences through the Containment Event Tree (CET) and determining the radionuclide releases for the various pathways through the CETs. To make this process more manageable, core damage sequences with similar characteristics are grouped into Plant Damage States (PDSs). This grouping procedure was developed through an iterative process resulting in a method that allowed core damage sequences to be grouped according to the status of plant systems at the onset of core damage (Duke 1990). PDSs are a combination of three separate binning characteristics:  1. Core melt bin - describes the status of the primary (reactor coolant) system and related systems during core damage. 2. Containment safeguards state - describes the status of containment related systems. 3. Containment isolation state - determines whether or not containment is isolated. The description of the binning process is discussed in terms of assigning sequences to core melt bins and use of a "bridge" tree to categorize containment safeguards/isolation states; however, these are concepts that are applied in the nodal logic of the CET rather than complete, stand alone event trees or decision trees. For example, each CET sequence includes all core damage cutsets in the "initiating event" of the sequence, but for each node, specific core melt
 
binning logic is used to quantify the node. For the "BYPASS" CET node, one of the inputs is a gate containing Core Melt Bin 19 events (CM-019), which are ISLOCA events. Gate CM-019 was manually created based on the Plant Damage State rules and used for the "BYPASS" evaluation because it satisfied the requirements for "containment bypass" cases. Similarly, the containment safeguards/isolation state bridge tree was used to manually develop logic gates for use in the CET nodes. An example of how the bridge tree logic is used is the evaluation of the "Fission Product Scrubbing is Effective" node. One potential means of scrubbing is the "plateout" mechanism, which is possible for releases that occur in the lower Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-9 License Renewal Application section of the auxiliary building. These include safeguard/isolation states G through R of the bridge tree. Logic representing these safeguard/isolation states was developed and included in the CET logic to allow only those sequence including isolation failures to pass through "plateout" logic for the "Fission Product Scrubbing is Effective" node.
Section E.2.2.3.1.1 provides further details rel a t ed to the d e velopment o f the PDS definitions for TMI-1. E.2.2.3.1.1 DEVELOPMENT OF PLANT DAMAGE STATES The plant damage states consider both the characteristics of the core material released to the environment and the mechanism by which the release is made from the containment. The content of the release is determined by the multiple factors, including the way in which core debris interacts with the containment and on the operation of mitigating systems, such as containment spray. The containment failure mode determines other factors such as the size and timing of the release. These issues are described in more detail in the following subsections.
E.2.2.3.1.1.1 SEQUENCE CHARACTERISTICS THAT AFFECT CONSEQUENCE ANALYSIS Source Term Magnitude and Isotopic Content The magnitude and isotopic content of the source term are affected by:
* The mechanisms by which radionuclides are released from the fuel,
* Retention of radionuclides in the primary system,
* The performance of active radionuclide removal systems such as the containment sprays,
* The mechanisms by which radionuclides are naturally removed from the containment atmosphere, and
* The mode of containment failure. The mechanisms by which radionuclides are released from the fuel depend on the progression of the accident. For example, if energetic attack of the concrete basemat by the core debris occurs, this can release large amounts of tellurium, a significant contributor to early fatalities. If Environmental Report Appendix E  SAMA ANALYSIS Page E-10 Three Mile Island Nuclear Station Unit 1 License Renewal Application a continuous supply of water contacts the core debris, a coolable debris bed can be formed and the tellurium release can be prevented or terminat ed (FAI 1987). Thus, it is necessary to know what plant conditions cause water to be present in the reactor cavity and at what times. Retention of fission products in the primary system can also be affected by system response. For example, core melt sequences following a large LOCA would result in significantly less primary inventory retention than would station blackout core melt sequences. Additionally, such factors as secondary side heat removal (SSHR) also affect the likelihood of revaporization of deposited radionuclides later in an accident. Revaporization of deposited radionuclides near the time of containment failure can significantly increase the release to the environment from a late containment overpressurization. Active radionuclide removal is accomplished by the containment sprays (NRC 1982). Containment sprays affect the magnitude of the source term by removing radionuclides from the atmosphere. Sprays affect the isotopic content of the source term because they are much more efficient in removing particulates than other forms of radionuclides. Therefore, it is necessary to know if and when containment sprays are operating. Natural removal processes also affect the magnitude of the source term. The effectiveness of gravitational settling and plateout on walls is dependent to a certain extent on the thermal-hydraulic conditions of the containment atmosphere. More importantly, it depends on the residence time of radionuclides in a given volume and thus on the type and time of containment
 
failure. Containment Failure The energy and duration of the radionuclide release and the warning time for evacuation are influenced by the type and time of containment failure. A structural (large breach) failure due to overpressurization will have a high energy of release as the containment rapidly depressurizes to atmospheric pressure from its failure pressure. The duration of release will be short due to the rapidity of the depressurization. Containment leakage due to an isolation failure would be more gradual. The duration of the release would be longer, and the energy associated with that release would be lower than for the puff release from overpressurization-induced failure. If containment integrity is maintained and the only releases are associated with design leakage, the energy of release is negligible and its duration is very long. Thus, the energy and duration of release depends on the type, or mode, of containment failure.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-11 License Renewal Application Warning time for evacuation is the time between the loss of long-term cooling capability and the release of radioactivity to the environment. An early core melt followed by an early containment failure (prior to 5 hours) does not allow much warning time (approximately 0-2 hours), whereas a late overpressurization may be gradual and predictable, allowing a significant amount of time for evacuation. The timing of containment failure can thus have an effect on warning time. Containment overpressurization can result from large combustible gas burns, steam spikes, direct containment heating (DCH), and a gradual buildup of steam and/or non-condensables. Since TMI-1's containment is constructed on limestone concrete, core-concrete interaction results in significant non-condensable gas (e.g., CO and CO2) production. Carbon monoxide is a combustible gas. The computer code MAAP, which was used to model containment behavior following postulated core melt events, allows for carbon monoxide to burn in the same fashion as hydrogen for combustible gas burns in containment. Combustible gas burns are influenced by the concentrations of oxygen and steam within the containment. The timing and severity of a combustible gas burn can also depend on the rate at which hydrogen is released to the containment from the Reactor Coolant System (RCS). In general, the larger the leak path (break size), the faster the hydrogen is released and the smaller the amount that is retained in the RCS until reactor vessel failure. The leakage path also affects the rate of hydrogen production in the core by controlling the release rate of steam.
The amount of steam available for the oxidation reaction affects the rate at which hydrogen is formed. The Containment Air Cooling Units (CACUs) also affect the combustible gas phenomenon within the containment. The CACUs are responsible for removing heat from the containment atmosphere and they also circulate the air within the containment, thus developing uniform concentrations of atmospheric constituents. The CACUs reduce the steam concentration, thus providing more suitable conditions for combustible gases to burn. However, the operation of the CACUs will also lower the containment base pressure and help to mitigate the effects of a combustible gas burn.
DCH is another phenomenon that can lead to co ntainment overpressurization. This phenomenon is important for sequences in which a core melt is initiated while the RCS is at a high pressure. It has been hypothesized that the corium (molten core material) can be ejected, under high pressure, from the reactor vessel and be dispersed into the containment atmosphere as finely fragmented particles. Airborne particulate debris could then rapidly release chemical (oxidation of metallic constituents) and thermal energy directly to the containment atmosphere.
Environmental Report Appendix E  SAMA ANALYSIS Page E-12 Three Mile Island Nuclear Station Unit 1 License Renewal Application Although the CACUs are not sufficient to stop DCH from occurring, their operation would be expected to lower the containment base pressure and thus help to mitigate the effects of DCH.
The containment sprays can also help mitigate the effect of combustible gas burns and DCH by reducing the static containment pressure. It has been stated that the warning time for evacuation is defined as the time between loss of long-term cooling capability and the release of radionuclides to the environment. The time from shutdown to the loss of long-term cooling capabilit y impacts the warning time given that the core decay heat load, and therefore the time to core melt, is a function of time. Even though recommended evacuation times are much longer than two hours, studies of past evacuations have shown that two hours is more than sufficient time to evacuate the majority of the population participating in the evacuation plan (PRC 1981). The SAMA evaluation uses site specific analysis to evaluate the impact of evacuation of offsite consequence, as described in Section E.3.6. E.2.2.3.1.1.2 CORE MELT BINS The core melt bin is the first of three characteristics that define the PDS. The core melt bin definition describes the status of the RCS and associated systems at the onset of core damage. 
 
Table E.2-2 lists the 19 core melt bins used in t h e TMI-1 PRA and provides a brief d efinition of each. This section describes the derivation of the core melt bin definitions, in terms of the RCS
 
leakage rat e , loss of p r i m ary s y stem makeup capability, a n d the condit i on of SSHR.
Tables E.2.3 through E.2.13 document how each of the core damage sequenc e s are assig n ed to the core melt bins. E.2.2.3.1.1.2.1 Reactor Coolant System Leakage Rate The RCS leakage rate is important in binning core damage sequences because it affects primary system pressure, timing of core damage, fission product retention in the primary system, and hydrogen release rate. There are four distinct leakage rate categories:
* Small LOCA,
* Medium LOCA,
* Large LOCA, and Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-13 License Renewal Application
* Cycling relief valve. Also, there are two special leakage categories:
* Steam generator tube rupture (SGTR),
* Interfacing systems LOCA. The small LOCA leakage rate is small enough that SSHR is effective in delaying core damage. Also, for small LOCAs, the primary system pressure will remain high during core damage (expected pressures are in the 1000 psia range) and may lead to a high-pressure melt ejection (HPME) when the reactor vessel fails. For TMI-1, a core melt sequence can be grouped as a small LOCA if it has one of the following break size characteristics:
* 0.007 ft2 to 0.1 ft2 breaks (DE&S 1992)
* Stuck open pressurizer PORV
* Stuck open SRV
* Reactor coolant pump (RCP) seal LOCA
* Steam generator tube ruptures that have an intact secondary system The medium LOCA leakage rate is a primary system failure that is small enough that the primary pressure will be relatively low (expected pressures are in the 300 to 400 psia range) so that the risk of a HPME accident is significantly reduced. For TMI-1, the medium LOCA size is from 0.1 ft2 to 0.5 ft2 breaks (DE&S 1992). The large LOCA leakage rate is large enough that the primary system pressure will be very low (expected pressure is less than 200 psia) so that there is little risk of a HPME. For TMI-1, breaks of this size are equal to or larger than 0.5 ft2. For those primary system ruptures involving pressurized thermal shock (PTS), a rapid cooling transient stress on the reactor vessel while at relatively high pressure, it was assumed that for this condition to occur, some type of primary injection must have been successful in order to achieve the requisite low temperatures at pressure. It was further assumed that the PTS condition would lead to a rupture of a size equivalent to a large LOCA. Therefore, those core Environmental Report Appendix E  SAMA ANALYSIS Page E-14 Three Mile Island Nuclear Station Unit 1 License Renewal Application damage sequences identified by PTS were categorized as large LOCA with successful injection but failure of early recirculation, i.e., core melt bin 2. A stuck open or cycling pressurizer relief or safety valve sequence would result in the primary system pressure remaining high (around the PORV set point) such that, if core melt occurred, the risk of a HPME would be high. In general, non-LOCA core melt sequences, such as transient and loss of offsite power (LOOP) sequences, were grouped with the cycling relief valve core melt category. The leakage category SGTR represents those steam generator tube rupture sequences where there is also a failure of the secondary system. This would result in a direct path for fission product release to the environment with little or no possibility of retention. The scrubbing and retention that is provided by SGTRs with intact steam generators are sufficient to group these with the intact containment plant damage states. For event sequences involving an intact primary system, i.e., no LOCA, and only a tube rupture within a single generator (with failure of
 
the second a ry system), core melt bin 16 (Table E.2-2) was chosen to re present this p articular
 
scenario.
The interfacing systems LOCA leakage category contains core melt sequences resulting from a rupture in a low-pressure system connected to the primary system. These sequences result in fission product releases that bypass the Reactor Building, but there is still the possibility of some retention in the buildings outside containment. E.2.2.3.1.1.2.2 Loss of Primary System Makeup Capability For core damage to occur, multiple failure of mitigation systems must occur. The timing and mode of failure of the primary system makeup capability can affect the characteristics of the PDS. For example, for sequences involving a loss of primary coolant, the timing of core damage is significantly affected by the time at which the safety injection systems fail. Also, the status of safety systems helps to determine whether or not the reactor cavity can be flooded, which impacts the post-core damage analysis. Core damage sequences are grouped into one of the following groups:
* Injection failure - sequences in which injection systems fail initially and do not inject the Borated Waste Storage Tank (BWST) contents into containment.
* Recirculation switchover failure - sequences in which injection systems fail when the BWST Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-15 License Renewal Application contents have been injected into containment and switchover to sump recirculation is attempted.
* Recirculation run failure - sequences in which injection systems switchover to sump recirculation following successful injection, but then fail later due to a run failure of the injection or support systems. A simplification was made with regard to start and run failures associated with recirculation of water from the containment sump. The dominant early recirculation failures (start failures) were associated with either failure of human actions (including dependent actions) involving operator switchover to sump recirculation prior to emptying the BWST or common cause failure mechanisms, such as valves DH-V-6A, and -6B, or DH-7A, and -7B failing to open, or the DHR pumps both failing to start. All other failures were assumed to be non-dominant start failures or are those that are truly designated as run failures, e.g., heat exchangers plugging, valves failing
 
to remain open, etc. E.2.2.3.1.1.2.3 Condition of Secondary Side Heat Removal The status of the SSHR System at the onset of core damage is also an important characteristic of the core damage sequence. SSHR can affect the time of core damage, the primary system pressure, the fission product retention in the primary, and operator actions that might affect core damage progression. There are two categories for the SSHR status:
* SSHR is available.
* SSHR is unavailable.
E.2.2.3.1.1.2.4 Anticipate d Transients Without SCRAM
: 1. For Anticipated Transients without Scram (ATWS) scenarios, the reactor fails to trip in conjunction with another initiating event that prompted the trip signal. Failure to trip the reactor when a valid trip signal occurs results in excessive thermal energy increasing reactor coolant system (RCS) pressure and temperature. The assumptions given below were imposed in order to associate the various ATWS scenarios with the appropriate, or
 
at least co n s ervative, core melt definition from Table E.2-2. Also, since t he ATWS event tree (Exelon 2005a) did not address availability of high-pressure injection for certain sequences that lead directly to core damage, failure of early injection was assumed. 2. Core moderator temperature coefficient (MTC) provides a natural feedback control mechanism in which core power is reduced as the moderator (reactor coolant) temperature increases. It is a function of the time in cycle for a given core. For Environmental Report Appendix E  SAMA ANALYSIS Page E-16 Three Mile Island Nuclear Station Unit 1 License Renewal Application conditions where there is an unsatisfactory moderator temperature coefficient (e.g., early in a cycle), excessively high RCS pressures may result due to insufficient negative feedback. Although the precise impact on the RCS in such a scenario depends on many other factors and could be benign, this scenario was assumed to result in a large LOCA without the ability to use high-pressure injection (core melt bin 1). 3. For both pressurizer safety valves and PORV unavailability, a large LOCA scenario without high-pressure injection (core melt bin 1) was assumed. 4. For loss of feedwater and inadequate secondary side pressure relief, the core melt bin with cycling primary relief valve without injection (core melt bin 12) was assumed. 5. For those scenarios with explicit failure of high-pressure injection, core melt bin 12 was assumed. SSHR, even if successful, was assumed inadequate for RCS heat removal. 6. For those scenarios involving failure of high-pressure recirculation, core melt bin 14 was assumed (late recirculation failure) instead of bin 13, since it is not clear that this sequence would actually lead to core damage.
E.2.2.3.1.1.3 CONTAINMENT SAFEGUARDS STATES The containment safeguards state is the second of three characteristics that define the PDS. The containment safeguards state describes the status at the onset of core damage for systems that provide a containment protective function. These systems include the reactor building spray system, and the containment air cooling units (CACUs), which are part of the reactor building emergency cooling system. These systems affect many decisions in the CET and, as a result, they affect accident progression. For example, the containment sprays affect fission product scrubbing, the flooding of the reactor cavity, and the time to reach core damage. E.2.2.3.1.1.4 CONTAINMENT ISOLATION STATES The third and final PDS characteristic is the status of containment isolation. Containment isolation is critical to preventing fission product release to the environment. Scoping studies with the MAAP computer code have indicated that there are two categories of containment isolation failure. Small isolation failures allow fission product releases much greater than those for an isolated containment with design leakage. However, the small isolation failures are less severe than early containment failures because they significantly reduce the release rate of fission products. Large isolation failures provide little or no delay in the fission product releases and are essentially the same as early containment failures. For TMI-1, the three containment isolation states are:  1. Isolated - containment is properly isolated.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-17 License Renewal Application
: 2. Small isolation failure - containment failure prior to core damage with hole size less than or equal to six inches. A small isolation failure precludes late overpressurization of containment, but does not preclude early overpressurization of containment. 3. Large isolation failure - containment failure prior to core damage with hole sizes larger than six inches. Large isolation failures preclude both early and late overpressurization of containment. E.2.2.3.1.1.5 PLANT DAMAGE STATE DEFINITION The PDSs are developed by combining the core melt bins, the containment safeguards states, and the containment isolation states. The PDS definition contains sufficient information about the sequences grouped into it that they may be treated as one. This information is critical input information for solving the CET. The PDS, rather than the individual sequences, determine the branch point frequencies in the CET.
Table E.2-2 lists and describes the 1 9 core melt bins and T a ble E.2-14 lists and describes the 18 combinations of containment safeguards states and containment isolation states. PDSs are described by a two-designator variable as follows:
XY Where:  X = core melt bin  Y = containment safeguards and isolation states The containment safeguards and isolation states were determined by the use of an event tree termed the "bridge" tree, since it bridges the gap between core melt scenarios, plant damage states, and the containment event tree for quantification of release categories. As discussed in
 
Section 2.2
.3.1 , these P D Ss manif e st themselves in the model as l o cal portions of CET nodal logic representing the applications of the binning concepts described above. There are no PDS flags associated with the cutsets as is common in other Level 2 model applications. E.2.2.3.2  Containment Event Tree Purpose The purpose of the Containment Event Tree (CET) is to quantify containment failure modes and radionuclide releases. Any phenomena that have a significant effect on the radionuclide release fractions or the timing, energy, and duration of the release are included in the tree as a top (header) event. The core damage sequences were categorized into Plant Damage States (PDSs), as determined in Section 2.2.3.1. These core damage sequences are trea t ed as Environmental Report Appendix E  SAMA ANALYSIS Page E-18 Three Mile Island Nuclear Station Unit 1 License Renewal Application initiating events for the CET. The paths that the PDSs can take through the event tree depend on how they affect the various events modeled. Because the path taken at each top event is based on probabilities and system fault tree evaluations, each PDS will appear at more than one CET end point with varying frequency. Thus, each end point can have more than one PDS state contributing to its total frequency. E.2.2.3.2.1 CONTAINMENT EVENT TREE DESCRIPTION Containment event trees, in some cases, have become so complex that the CETs can not be easily represented and are difficult to understand by anyone other than a consequence analyst. The approach used for the TMI-1 analysis relies on converting the large and complex CET into a combination of a small event tree and large decision trees. In developing the TMI-1 small CET, the only questions included are those that have an effect on the release timing, energy, location, or fission product fractions. When completed, each CET end state represented a separate release category. The CET release category results are presented in Section 2.2
.3.3. After the containment event tree was developed, decision trees using both success and failure logic were developed to determine the probability of the appropriate top event (node) in the CET. This approach was used to avoid the use of NOT gates for sequence success logic, which tended to make the model more complicated and difficult to quantify. The CET developed for TMI-1 consists of 11 nodal top events that were modeled via the use of Boolean logic, for both success and failure of each branch. The following section defines and describes the CET top events and their associated decision trees. The top events are summarized in Table E.2-15. The logic for each of the CET nodes is cumbersome and complex, so it is not included in this discussion. To make use of the CET, the important characteristics of the plant's containment must be identified. Three of the more important features that must be considered are the containment ultimate strength capacity, the concrete type, and the reactor cavity arrangement. The ultimate capacity of containment provides the basis for establishing containment failure probability and failure modes given various accident progression scenarios. TMI-1, is a Babcock & Wilcox PWR with vertical straight-tube (once-through) steam generators that produce superheated steam at constant pressure. The reactor and the nuclear steam supply Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-19 License Renewal Application system are contained within a Reactor Building that is a post-tensioned reinforced-concrete cylinder and dome. The interior of the surface of the building is lined with a one-quarter inch thick welded steel plate to ensure a high degree of leak tightness. Generally, TMI-1 can be placed into the category of PWR large dry containments, because of their high mean failure pressure, overall containment volume, and open lower containment configuration. The type of concrete affects the type and properties of gases released during concrete attack. TMI-1's concrete contains a limestone aggregate, which can result in significant non-condensable gas production during concrete ablation. The reactor cavity geometry affects how (or if) water can reach the cavity during a core damage sequence. The cavity arrangement is important when considering the following phenomena:
* Ex-vessel debris bed coolability
* Potential for direct containment heating
* Ex-vessel steam generation
* Ex-vessel hydrogen or combustible gas production
* Ex-vessel fission product release
* Hydrogen or combustible gas recombination
* Long-term containment overpressurization
* Basemat melt-through
* Potential for debris-liner contact
* Sources of water and pathways to the lower reactor cavity E.2.2.3.2.2 CONTAINMENT EVENT TREE TOP EVENTS In this section, the CET top events are defined and described. The CET top events are summarized in Table E.2-15.
Environmental Report Appendix E  SAMA ANALYSIS Page E-20 Three Mile Island Nuclear Station Unit 1 License Renewal Application CET Top Events Description A: Containment Bypass Does the release of radionuclides take place within the containment? Success for this event means that containment is available as a barrier to fission product release. Failure means containment is not available as a barrier to fission product release. The types of accidents that bypass the containment are steam generator tube ruptures (as an initiating event or an induced event) and interfacing-systems LOCA. This top event is further developed using a decision tree model. B: Containment Isolation Does the containment isolate such that: 1) a leakage rate sufficient to cause a substantial increase in radionuclide release to the environment does not occur, and 2) containment pressure response is not significantly affected? Success for this event means that containment isolation performs its function so that containment becomes a barrier against flow of radionuclides to the environment. Failure means containment integrity is lost and a path is available for radionuclides to reach the environment. This event is concerned with the time at the beginning of the accident sequence (i.e., when isolation occurs) before radionuclides are released to the containment atmosphere. C: Isolation Failure Size Is the isolation failure equivalent to a small hole size in containment? Success for this event means that the isolation failure is small, i.e., system top event SMALL-ISO. For the TMI-1 analysis, a small isolation failure is defined as a six-inch equivalent diameter hole. Isolation failures of this type allow some time for holdup inside containment where natural removal mechanisms (e.g., plateout) will reduce radionuclide concentrations. Failure of this event implies that the isolation failure is not small, i.e., system top event LARGE-ISO, and allows little or no holdup in containment. Both small and large isolation failures preclude late overpressurization. All other containment overpressure sequences (hydrogen burns, direct containment heating, etc.) are prevented only by large isolation failures. D: Auxiliary Building Release Does the fission product release pass through the Auxiliary Building? Success for this event means that the fission product re lease will pass through the Auxiliary Building. This release path is the result of an interfacing-system LOCA or an isolation failure to the Auxiliary Building. Failure for this event means that the fission product release does not pass through the Auxiliary Building. A release path that bypasses the Auxiliary Building is a pa thway directly to the environment. This top event is applicable only if containment is not isolated or is bypassed. Determination of success or failure depends on the type of isolation failure, where the fission products are released, and the PDS. For example, a SGTR would be a failure, while most interfacing systems LOCAs would be a success.
E: Early Containment
 
Failure Does the containment remain intact until long after reactor vessel failure (i.e., a time period which allow sufficient time for fission product settling)? Success for this event means that containment remains intact long after reactor vessel failure. Failure for this event means that containment has failed prior to or within the time required for fission product settling and decay of short-lived isotopes. This time period is typically defined as five hours after reactor vessel failure.
F: Late Containment
 
Failure Does the containment remain intact throughout the entire core melt sequence? Event success means that the containment remains intact throughout the entire core melt sequence. Releases to the environment after this point, if any, are due to normal containment leakage or basemat melt-through. Failure of this event means that containment fails late in the core melt sequence due to an overpressurization event.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-21 License Renewal Application CET Top Events Description G: Benign Containment Failure Is late containment failure benign?
Success for this event means that a late overpressurization results in a benign containment failure, i.e., leak-before-break. This failure mode is described as a series of small cracks that develop in the containment structure such that further pressurization does not occur. Failure of this event means that a late overpressurization results in a catastrophic containment failure, which would cause containment to depressurize rapidly. This is strictly a function of the containment type, and is quantified identically for all PDSs. H: Ex-Vessel Release Of Fission Products Is a coolable debris bed established outside the reactor vessel so that significant ex-vessel fission product releases do not occur? Success for this event means that a coolable debris bed is established in the reactor cavity or the containment, preventing an ex-vessel release. Failure means that a coolable debris bed is not established, allowing the corium to attack the concrete (producing non-condensable gases) and resulting in an ex-vessel release. The ex-vessel release involves a significant amount of tellurium and other fission products. I: Containment Basemat Failure Is a coolable debris bed established in the reactor cavity to prevent containment failure from basemat melt-through? Success for this event means that the debris bed in the cavity is cooled, and concrete ablation is stopped. Failure means that the debris bed is not cooled and ablates concrete until the basemat is failed. J: Revaporization Release Is a revaporization release of volatile fission products at or near the time of containment failure prevented? Success for this event means that large amounts of volatile fission products have not revaporized and are not available for release when containment overpressurizes. Failure means that volatile fission products that were deposited in the RCS have revaporized and are available to be released in large amounts when containment fails. Revaporization is only considered for late catastrophic containment failures. Early containment failures release fission products at or shortly after reactor vessel failure resulting in high release fractions. The effects of revaporization, if any, would not be seen for this failure mode. Late containment failures, however, provide time for radionuclide removal from the atmosphere by various methods. As a result, release fractions at containment failu re are lower so that revaporization of fission products will have a larger impact. Revaporization is not considered for benign failures of containment since the pressure remains high due to the slow depressurization of containment. Since the pressure remains high in containment, revaporization is
 
unlikely to occur. K: Fission Product Scrubbing Are fission product removal mechanisms available to reduce the amount of radionuclides released to the environment? Success for this event means that the fission products are scrubbed by some method prior to release to the environment. These mechanisms include:  - Containment scrubbing (e.g., sprays) - Auxiliary Building scrubbing (e.g., plateout) - Steam Generator scrubbing (e.g., water pool release) Failure for this event means the fission products are not scrubbed prior to release to the environment by any method.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-22 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.2.2.3.3 Release Categories and Source Terms The endpoint of the CET contains two major pieces of information, which are the release frequency and the release category designation. The parameters that define a release category and are important in the analysis of offsite consequences are:
: 1. T ime of release 2. Duration of release 3. Energy of release 4. Warning time for evacuation 5. Isotopic fractions released to the environment Each CET end point is capable of describing a unique sequence with potentially unique release characteristics. For TMI-1, 39 release categories were identified in the CET with most endpoints having a unique release category designation. A numbering scheme is used to separate major categories:
* 1 = Containment Bypass with Auxiliary Building Bypass
* 2 = Interfacing-Systems LOCA
* 3 = Large Isolation Failures
* 4 = Small Isolation Failures
* 5 = Early Containment Failure
* 6 = Late Containment Failure (Catastrophic)
* 7 = Late Containment Failure (Benign)
* 8 = Basemat Melt-Through
* 9 = No Containment Failure Different sequences within these major categories were given a designation such as 1.01, 1.02, 2.01, etc. in order to distinguish between specific details of the containment response. The 39 TMI-1 release categori e s are summarized in T a ble E.2-16.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-23 License Renewal Application The MAAP thermal hydraulics code was used to analyze the plant specific containment responses for each of the CET sequences. The 39 TMI-1 release categories were then reviewed in order to determine how they could be grouped for the assignment of source terms.
It is possible to develop source terms for every release category in the CET, but in many cases, the results are so similar that maintaining unique source terms for every release category does not provide any measurable benefit. As a result, release categories with similar traits were grouped together and a single source term was used to represent the entire group to streamline the Level 3 analysis. For TMI-1, nine major source term groups identified above were found to be an adequate structure for segregating the source terms. The table below correlates the major source term groups to the source term designators and provides basic descriptions of the representative sequence established for each source term group: Representative Sequence Descriptions for Source Term Groups Release Category Group Source Term Designator General Description of Contributing Sequences 1: Containment Bypass w/ Aux Bldg Bypass SGTR This event is initiated with a double ended failure of a steam generator tube with the SG safety valve failed open. All injection is assumed unavailable. Emergency feedwater is available.
2: ISLOCA ISLOCA This event is initiated with a small break outside of containment followed by failure of injection. Emergency feedwater is available. 3: Large Isolation
 
Failure ISO-LG This scenario is represented by a loss of main feedwater followed by a failure of all injection. A large containment isolation failure is assumed to occur at time zero. Emergency feedwater operates successfully for a period of 6 hours. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop. Core damage occurs at 9.4 hours into the event followed by failure of the hot leg due to creep rupture 36 minutes later. Vessel breach occurs at 16 hrs. 4: Small Isolation
 
Failure ISO-SM This scenario is represented by a loss of main feedwater followed by a failure of all injection. A small containment isolation failure is assumed to occur at time zero. Emergency feedwater is assumed unavailable. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop. Core damage occurs at 50 minutes into the event followed by failure of the hot leg due to creep rupture 36 minutes later. Vessel breach occurs at 6 hrs. 5: Early Containment
 
Failure EARLY This scenario is represented by a Station Blackout. Emergency feedwater operates successfully for a period of 6 hours. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop.
Core damage occurs at 9 hours into the event. Vessel breach occurs at 11.7 hrs. It is assumed that containment failure occurs at the time of vessel breach.
Environmental Report Appendix E  SAMA ANALYSIS Page E-24 Three Mile Island Nuclear Station Unit 1 License Renewal Application Representative Sequence Descriptions for Source Term Groups Release Category Group Source Term Designator General Description of Contributing Sequences 6: Late Containment Failure (catastrophic) LATE-LG This scenario is represented by a loss of main feedwater followed by a failure of all injection. Emergency feedwater operates successfully. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop. Core damage occurs at 26 hours into the event followed by failure of the hot leg due to creep rupture 40 minutes later. Vessel breach occurs at 34.8 hrs. The containment fails due to overpressure at 70 hours into the event with an assumed large failure area, resulting a rapid depressurization of containment. 7: Late Containment
 
Failure (benign) LATE-SM This scenario is represented by a Station Blackout. Emergency feedwater operates successfully for a period of 6 hours. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop.
Core damage occurs at 9 hours into the event followed by failure of the hot leg due to creep rupture 50 minutes later. Vessel breach occurs at 16.5 hrs. Containment sprays are assumed to be recovered at 24 hours into the event. The core debris remains covered with water, however, without heat removal, the containment fails due to overpressure at 52 hours into the event. The breach area is assumed to be represented by a leak-before-break and results in a very slow containment depressurization. 8: Basemat Melt-Through BMMT This scenario is represented by a loss of main feedwater followed by a failure of all injection. Emergency feedwater operates successfully. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop. Core damage occurs at 26 hours into the event followed by failure of the hot leg due to creep rupture 40 minutes later. Vessel breach occurs at 34.7 hrs. All of the core debris is forced to remain in the reactor cavity in order to accelerate the amount of core concrete attack. When concrete erosion has exceeded 6 feet, containment failure is assumed to occur with a representative failure area equal to 1 ft2. 9: No Containment
 
Failure INTACT This scenario is represented by a loss of main feedwater followed by a failure of all injection. Emergency feedwater operates successfully. At 15 minutes into the event, 42 gpm seal leakage is assumed per loop. Core damage occurs at 26 hours into the event followed by failure of the hot leg due to creep rupture 40 minutes later. Vessel breach occurs at 34.6 hrs. Successful operation of containment sprays and fan coolers prevents containment overpressure failure long term.
Table E.2-17 provides additional a c cident progr e ssion in f ormation for the representative sequences described above, including the time to core damage, time to containment failure, and notable release fractions.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-25 License Renewal Application In some cases, there were competing contributors to a release category group with measurable differences in some of the release fractions (e.g., scrubbed vs. unscrubbed releases). The representative source term for the release category is typically chosen based on the largest frequency, but when the consequences of a source term with a smaller frequency are more severe, the more severe source term is used if it is believed that the group would otherwise be underrepresented. The source terms that are used as input to the TMI-1 Level 3 model are a combination of radionuclide release fractions, the timing of the radionuclide release relative to the declaration of a general emergency, and the frequencies at which the releases occur. This combination of information is used in conjunction with other TMI-1 site characteristics in the Level 3 model to evaluate the conseque n ces of a co r e damage eve n t. Table E.2-18 provides a summary of the TMI-1 source term information, which includes the following:
* MAAP case identifier (for reference),
* Airborne release for each of the fission product groups provided my MAAP,
* Start time of the airborne release (measured from the time of accident initiation),
* End time of the airborne release (measured from the time of accident initiation).
Note that the individual release cat e gory frequencies are p r ovided in Table E.2-16. E.2.2.4 TMI MODEL RESULTS Figure E.2-1 is a pie-ch art showing t he initiati n g event contribution to in t ernal events CDF from the quantification of the TMI PRA 2004 Revision 2 model at a truncation limit of 1E-11.
Table E.2-19 presents the ra n k ed list of initiating events by their c o ntribution to CDF. As can be seen in the table, about a third of the total CDF comes from loss of offsite power events. About one-half of CDF is due to a combination of transients and very small break (<1.0" diameter) and small break LOCAs (1"-4.3" in diameter). The next largest single contributor is loss of nuclear services river water, which accounts for about 16% of CDF. It is interesting to note that the large LOCA initiator, which represents the design basis accident for TMI-1, accounts for less
 
than 1% of t he total CDF.
Figure E.2-2 is a bar chart displ a ying the system importa n c e rankin g s (basically by Fussell-Vesely). Onsite emergency electrical power and offsite power sources dominate the contributions to CDF.
Environmental Report Appendix E  SAMA ANALYSIS Page E-26 Three Mile Island Nuclear Station Unit 1 License Renewal Application The TMI PRA includes a Level 2 model from which each of the release category frequencies can be calculated. The Release Category results are based on the TMI 2004 Revision 2 model, which was completed in 2007.
Table E.2-20 presents the top initiating ev e nts for each of the release categories. With regard to Large Early Release frequencies (LERF), the TMI LERF is estimated at 3.0E-6/year (12.7% of CDF). These results are slightly higher when compared to other PWRs with large dry containments that generally fall in the range from 3% to 10% of CDF. The contributions to LERF consist of the following release categories:  RC1-02 RC3-03 RC4-04 RC2-01 RC3-04 RC4-05 RC2-02 RC3-05 RC4-06 RC2-03 RC3-06 RC4-07 RC2-04 RC4-01 RC4-08 RC3-01 RC4-02 RC5-01 RC3-02 RC4-03 RC5-02 E.2.3 EXTERNAL FLOODING MODEL The External Flooding model developed for the IPEEE was a simplified, Level 1 PRA evaluation. While there are words in the IPEEE that indicate it is a Level 2 analysis, the depth of any containment performance analysis that was carried out was not robust enough to support the SAMA analysis. In order to provide a means of evaluating the external flooding based SAMAs, it was necessary to develop representative source terms and release frequencies for the most important floo d ing contribu t ors. This p r ocess is described in S e ctions E.2.3
.1 and E.2.3.2. E.2.3.1 CORE DAMAGE SEQ UENCE IDENTIFICATION The core damage sequences developed for the external flooding model include three major
 
groups:
* Floods with elevations greater than 310 feet mean sea level (msl)
* Floods with elevations between 305 and 310 feet msl,
* Floods with elevations less than 305 feet msl.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-27 License Renewal Application Of these groups, the floods above 310 feet and those below 305 feet are each represented by a single core damage sequence. The floods between 305 and 310 feet are represented by six sequences that were quantified using an event tree developed specifically for the IPEEE external flooding evaluation. The descriptions and frequencies of these sequences are
 
summarized in Table E.2-21. E.2.3.2 LEVEL 2 BINNING OF EXTERNAL FLOODING SCENARIOS In order to provide the input required for the Level 3 analysis of the external flooding scenarios, it was necessary to use the information in the IPEEE to estimate the plant response after core damage. Two separate processes were required to address the different flood scenarios. For the 305' to 310' msl floods and the floods greater than 310' msl, the flooding sequences were analyzed and direct correlations between the core damage sequences and the source terms were developed. For floods below 305' msl, the containment performance characteristics for LOOP events were used to determine the releases given the similarity in the events.
E.2.3.2.1 Source Term Correlation for Ex ternal Flood Sequences Over 305' msl In order to determine the quantitative distribution of the flooding sequences among the TMI-1 source terms, it was necessary to make assumptions about the reactor status based on the information available in the IPEEE, determine which sequences should be binned to specific source terms, and then calculate the conditional probabilities of the relevant CET sequences. For cases where the transition to cold shutdown was not completed before accident initiation, a specific set of valves corresponding to a small pathway would be left open and a conditional probability of 1.0 was assigned to the "Iso Sm" source term (small isolation failure). These sequences are all from the IPEEE 305' to 310' msl flood cases and include:
* Sequence "B"
* Sequence "D"
* Sequence "E"
* Sequence "F" The remaining sequences are evolutions in which the plant is successfully transitioned to cold shutdown before the onset of accident conditions. These sequences include:
Environmental Report Appendix E  SAMA ANALYSIS Page E-28 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Floods >310' msl
* 305' to 310' msl flood sequences "A"
* 305' to 310' msl flood sequences "C" In these cases, there are a number of ways in which the containment could fail and the Level 2 CET was used to estimate the conditional failure probabilities assuming that containment isolation was initially successful. The conditional probabilities for theses sequences were calculated by quantifying specific nodal events in the CET that were chosen because they helped est a blish sou r ce term bins.
Table E.2-22 summarizes the binning characte ristics of each of these nodes:
A simplified version of the TMI-1 CET (see Figu r e E.2-3) has been developed using only these nodes to graphically depict the binning process and to document the fractional division of the relevant external flooding sequences among the source terms. Additional details related to the CET development and uses are provided in the TMI-1 Containment Event Tree Analysis
 
Notebook (Exelon 2007b).
E.2.3.2.2 Source Term Correlation for External Flood Sequences Below 305' msl External floods below 305' msl do not have an im pact on TMI-1 other than any LOOP event that may accompany the flood conditions, which is an insight that was used to estimate the
 
containment performance and release characteristics for these events. The PRA model was quantified with all initiating events other than LOOP set to zero in order to simulate the conditions expected to exist for external floods below 305' msl. The resulting release category frequencies were used to define the generic fractional distribution of these flood events among the 39 release categories. Review of the release category frequencies demonstrated that 95% of the risk is associated with only 8 of the release categories. In order to simplify external flooding calculations, only these 8 release categories are used in the external flooding quantifications. The 5.3 percent contribution from the non-used release categories has been accounted for by adding 5.31E-02 to the total for RC5-01, the "Early" release bin, which is conservative for the purposes of the SAMA analysis. The following table summarizes the RC fractions used in the quantifications:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-29 License Renewal Application RC name Probability Fraction of total Correction to account  for non-used RCs Revised RC fractions 1-02 1.59E-06 6.71E-02 0 6.71E-02 4-04 3.16E-07 1.33E-02 0 1.33E-02 5-01 7.39E-07 3.12E-02 5.31E-02 8.43E-02 7-03 7.45E-07 3.15E-02 0 3.15E-02 7-04 2.89E-07 1.22E-02 0 1.22E-02 8-01 3.19E-06 1.35E-01 0 1.35E-01 9-01 1.32E-05 5.57E-01 0 5.57E-01 9-03 2.36E-06 9.96E-02 0 9.96E-02 Totals  9.47E-01 5.31E-02 1.00E+00 The source t erms for these release categories a r e provided in Section E.2.2.3.3. E.2.3.2.3  External Flooding Binning Summary The desired product of the External Flooding binning process is a set of frequencies that are correlated to the TMI-1 source terms so that they can be used with the Level 3 model results to quantify the consequences of External Flooding accidents. The consequence results are then
 
used in the cost benefit analysis, as described in Section E.4. Table E.2-23 summar i zes the source term specific frequencies for each of the TMI-1 External Flooding sequences. E.2.4 TMI-1 PEER REVIEW
 
==SUMMARY==
 
The TMI-1 internal events PRA received a formal industry PRA Peer Review in August 2000. The final report was issued in March, 2001.  "It was the general assessment of the peer review team that the Three Mile Island PRA can be effectively used to support applications involving risk significant determinations supported by deterministic analysis, once the technical issues and recommendations for enhancements that are noted in the element summaries and Fact and Observation Sheets are addressed to an appropriate level of quality."
Table E.2-24 contains t he grades of the individ u al PRA Elements recorded by the Peer Revi e w Team.
All 'A' and 'B' F&Os are closed with exception of one 'B' level observation. F&O SY-21 relates to the need for independent technical and system engineer reviews of system notebooks. Most of the system notebooks have not been systematically reviewed by the system engineers.
Environmental Report Appendix E  SAMA ANALYSIS Page E-30 Three Mile Island Nuclear Station Unit 1 License Renewal Application The Peer Review Report also credits items of strength in the TMI PRA. Some of the strengths were:
* Treatment of dependencies in sequence and system models. There was excellent treatment and documentation of system functional dependencies and physical dependencies evidenced by system dependency matrices.
* Room heatup tests to support model. To resolve some earlier uncertainties regarding the impact of loss of room cooling to the electrical switchgear rooms and other areas, TMI conducted test to verify the success criteria for associated HVAC systems.
* Excellent ISLOCA treatment. The treatment of interfacing system LOCA sequences, including the systematic review of candidate pathways, quantification of initiating event frequencies, evaluation of response of low pressure systems to overpressure, and treatment of containment isolation interfaces, was state of the art and well conducted.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-31 License Renewal Application E.3 LEVEL 3 PRA ANALYSIS This section addresses the critical input parameters and analysis of the Level 3 portion of the probabilis tic risk assessment. In addition, Sect i on E.7.3 s u mmarizes a series of s e nsitivity evaluations to potentially critical parameters. E.3.1 ANALYSIS The MACCS2 code (NRC 1998a) is used to perform the Level 3 probabilistic risk assessment (PRA) for the Three Mile Island Nuclear Generating Plant. Three Mile Island site specific parameters are used for population distribution and economic parameters. Plant-specific release data included the time-dependent nuclide distribution of releases and release frequencies. The behavior of the population during a release (evacuation parameters) is based on plant and site-specific set points. Other input parameters given with the MACCS2 "Sample Problem A", formed the basis for the present analysis. These data are used in combination with site-specific meteorology to simulate the probability distribution of impact risks (both exposures and economic effects) to the surrounding 50-mile radius population as a result of the release accident sequences at Three Mile Island.
E.3.2 POPULATION The population surrounding the Three Mile Island site is estimated for the year 2034. Population projections within 50 miles of Three Mile Island are determined using SECPOP2000, (NRC 2003) utilizing a geographic information system (GIS). U.
S Census block-group level population data is allocated to each sector based on the area fraction of the census block-groups in that sector. U.S. Census data from 1990 and 2000 are used to determine a ten year population growth factor for each of the 50-mile radius rings. The population growth factor for each ring is applied uniformly to all sectors in the ring to calculate the year 2034 population distribution. The distribution is given in terms of population at distances to 1, 2, 3, 4, 5, 10, 20, 30, 40 and 50 miles from the plant and in the direction of each of the 16 compass points (i.e., N, NNE, NE--NNW). The total year 2034 population for the 160 sectors (10 distances x 16 directions) in the region is estimated as 3,609,252. The ten year population growth factor (in parenthesis) and distribution Environmental Report Appendix E  SAMA ANALYSIS Page E-32 Three Mile Island Nuclear Station Unit 1 License Renewal Application of the population is given for the 10-mile radius from Three Mile Island and for the 50-mile radius from Three Mile Island in Ta b les E.3-1 a n d E.3-2 , respectively.
E.3.3 ECONOMY MACCS2 requires certain economic data (fraction of land devoted to farming, annual farm sales, fraction of farm sales resulting from dairy production, and property value of farm and non-farm land) for each of the 160 sectors. These values are calculated using the SECPOP2000 code (NRC 2003). SECPOP2000 utilizes economic data from the U.S. Department of Agriculture, "1997 Census of Agriculture" (USDA 1998) and from other 1998 and 1999 data sources. Economic values for up to 97 economic zones are calculated and allocated to each of the 160 sectors. In addition, generic economic data that are applied to the region as a whole are revised from the MACCS2 sample problem input when better information is available. These revised parameters include per diem living expenses (applied to owners of interdicted properties and relocated populations), relocation costs (for owners of interdicted properties), and value of farm and non-farm wealth. These values are updated to the year 2006 value using the Consumer Price Index
 
ratio. Three Mile Island MACCS2 economic parameters include the following: Three Mile Island MACCS2 Economic Parameters Variable Description Three Mile Island Value DPRATE (1) Property depreciation rate (per yr) 0.2 DSRATE (1) Investment rate of return (per yr) 0.12 EVACST (2) Daily cost for a person who has been evacuated ($/person-day) 48.72 POPCST (2) Population relocation cost ($/person) 9022.00 RELCST (2) Daily cost for a person who is relocated ($/person-day) 48.72 CDFRM0 (2) Cost of farm decontamination for various levels of decontamination ($/hectare) 1015.00 2256.00 CDNFRM (2) Cost of non-farm decontamination per resident person for various levels of decontamination ($/person) 5413.00 14435.00 DLBCST (2) Average cost of decontamination labor  ($/man-year) 63155.00 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-33 License Renewal Application Three Mile Island MACCS2 Economic Parameters Variable Description Three Mile Island Value VALWF0 (3) Value of farm wealth ($/hectare) 3311.00 VALWNF (3) Value of non-farm wealth ($/person) 110473.00 (1) DPRATE and DSRATE are based on NUREG/CR-4551 value (NRC 1990).
(2) These parameters for Three Mile Island use the NUREG/CR-4551 value (NRC 1990), updated top the 2006 CPI value.
(3) VALWF0 and VALWNF are based on SECPOP2000 values for Three Mile Island, updated to the 2006 CPI value.
E.3.4 FOOD AND AGRICULTURE Food ingestion is modeled using the COMIDA2 methodology consistent with Sample Problem A. The COMIDA2 model utilizes national based food production parameters derived from the annual food consumption of an average individual such that site specific food production values are not utilized. The fraction of population dose due to food ingestion is typically small compared to other population dose sources. For Three Mile Island, approximately five percent of the total population dose is due to food ingestion. E.3.5 NUCLIDE RELEASE MACCS2 requires input for 60 radionuclides. The core inventory at the time of the accident is based on a plant specific ORIGEN2.1 calculation for a 24 month refueling cycle (obtained from C-1101-900-E-220-178, Rev. 0, 20 0 2). Table E.3-3 provides the MACCS2 Three Mile Island
 
core inventory. Three Mile Island nuclide release categories are related to the MACCS categories as shown in Table E.3-4. All releas e s are modeled as occur r ing at 51.6 meters (top of the Reactor Building).
The thermal content of each of the releases are assumed to be 1.0E+07 watts based on values provided in Sample Problem A and NUREG/CR-4551 (NRC 1990). The release associated with each source term is modeled as two or three individual plume segments to capture nuclide release changes as a function of time. Two nuclide release sensitivity cases were performed to determine the effect of release height and thermal content assumptions. One sensitivity case modeled the releases occurring at ground level (0.0 meters). The second sensitivity case modeled the thermal content of each Environmental Report Appendix E  SAMA ANALYSIS Page E-34 Three Mile Island Nuclear Station Unit 1 License Renewal Application release to be the same as ambient (i.e., buoyant plume rise is not modeled). The results are discussed in Section E.7.3. E.3.6 EVACUATION Reactor scram signal begins each evaluated accident sequence. A General Emergency is declared when plant conditions degrade to the point where it is judged that there is a credible risk to the public. Therefore, the timing of the General Emergency declaration is sequence specific and ranges from 48 minutes to 26 hours for the release sequences evaluated. The MACCS2 User's Guide input parameters of 95 percent of the population within 10 miles of the plant [Emergency Planning Zone (EPZ)] evacuating and 5 percent not evacuating are employed. These values have been used in similar studies (e.g., Hatch, Calvert Cliffs, (SNOC 2000) and (BGE 1998)) and are conservative relative to the NUREG-1150 study, which assumed evacuation of 99.5 percent of the population within the EPZ. The evacuees are assumed to begin evacuating 90 minutes after a General Emergency has been declared and are evacuated at an average radial speed of 1.18 miles per hour (0.53 m/sec). This speed is the time weighted value accounting for season, day of the week, time of day, weather conditions, and special events. The evacuation time weighted average of 600 minutes is for the full 0-10 mile EPZ, an assumed 15 minute notification time, 15 minutes for evacuation preparation, and 60 minutes average departure time. (ETI 2003) One evacuation sensitivity case is performed to determine the impact of evacuation assumptions. The sensitivity case reduced the evacuation speed by a factor of two (0.26
 
m/sec). The results are discussed in Section E.7.3. E.3.7 METEOROLOGY Annual Three Mile Island meteorology data from year 1998 is used in MACCS2 for the base case results. The year 1998 meteorological data set is utilized for the Three Mile Island base case MACCS2 analysis based on the fact that the year 1998 provided the most complete data set, the highest population dose risk and offsite economic cost risk, and is judged to be the most
 
conservative. Year 1998, 1999, and 2000 meteorology data for the Three Mile Island site contains wind speed, wind direction, and stability data. Site specific precipitation data was not included. The 1998 Three Mile Island meteorological data set contained 39 total hours of missing data, Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-35 License Renewal Application representing 0.45% of the hourly readings. The 1999 and 2000 Three Mile Island meteorological data sets contained 54 and 23 total hours of missing data, respectively, representing 0.62% and 0.26% of the hourly readings. Of the three data sets used the 1998 data set is the only data set that did not include any gaps of missing data of more than two hours. Therefore, it is judged the year 1998 provided the most complete data set. The year 1998 meteorological data set contained several one or two hour gaps of missing data (39 hours, 0.45%). Traditionally, up to 10% of missing data is considered acceptable. All of the missing gaps consisted of two hours or less and interpolation was used to fill in the missing meteorological data. It is noted that MACCS2 results used in the SAMA analysis are the statistical mean of 384 weather sequences (each sequence contains 120 hours of data) chosen at random from pre-sorted weather bins. Due to the large number of samples analyzed, the adjustment of any particular weather sequence has negligible impact on the mean results. Three Mile Island MACCS2 analysis evaluated three meteorological data sets (Calendar years 1998, 1999, and 2000) to ensure that the meteorological data set used in the analysis is adequate. The use of the most conservative data set (year 1998) accounts for any weather sequences that may have been misrepresented by substitute data. Based on the multiple years analyzed, minimum data gaps in the year 1998 meteorological data, and the sampling methodology used, the reported mean results are judged acceptable and appropriate for use in averted cost risk calculations. Meteorological data is prepared for MACCS2 input as follows: 1. Wind speed, wind direction, and atmospheric stability data is provided from the site. Precipitation data from the Middletown/Harrisburg Airport is utilized. 2. If a brief period (i.e., < 6 hr.) of missing data exists, interpolation is used between hours. 3. For larger data voids (i.e., > 6 hr.), data from the previous or following day is utilized to fill data gaps (for the same time of day). 4. Atmospheric mixing heights are specified for morning and afternoon. These values were taken from the document Mixing Heights, Windspeeds, and Potential for Urban Air Pollution throughout the Contiguous United States (EPA 1972).
 
This source defined morning as being the four-hour period from 0200 to 0600 Local Standard Time and afternoon as being the four-hour period from 1200 to 1600 Local Standard Time. 
 
The Code Manual for MACCS2: Vo l ume 1 (from Appendix A, pages A-1 and A-2) states Environmental Report Appendix E  SAMA ANALYSIS Page E-36 Three Mile Island Nuclear Station Unit 1 License Renewal Application the following:
"The first of these two values corresponds to the morning mixing height and the second to the afternoon height. In the current implementation, the larger of these two values and the value of
 
the boundary weather mixing height is used by the code." 
"In its present form, that atmospheric model implemented in MACCS2 does not allow a change in the mixing layer to occur during transport of the plume. Mixing layer height is assumed to be constant and therefore only a single value is used by the code." For the Three Mile Island MACCS2 analyses, these conditions mean that, only the afternoon mixing height is used since it is larger than the morning mixing height. Note that the boundary weather mixing height, wind speed and stability category are only used when there is no meteorological data. These fixed boundary weather values are ignored by the code when an hourly meteorological data file is supplied by the user, as was the case in the MACCS2 runs for
 
Three Mile Island. As noted above, site meteorological data for years 1999 and 2000 are also evaluated as sensitivity cases to ensure year 1998 data is an appropriate data set. The results are discussed in Section E.7.3. E.3.8 MACCS2 RESULTS Tables E.3
-5 shows the mean off-site doses and economic impacts to the region wi t hin 50 miles of Three Mile Island for each of nine source term groups evaluated using MACCS2. These impacts are multiplied by the annual frequency for each release category and then summed to obtain the dose-risk and offsite economic cost-risk (OECR) for the TMI-1 internal events initiators.
T able E.3-6 provides the results for t h e non-zero r elease cat e gories. Table E.3-7 summarizes the base c a se results f o r the seque n ce specific e x ternal flooding contributio ns based on t he source t e rm frequencies identi f ied in Section E.2.3.2.3 and the source term specific dose and cost r esults i d enti f ied in Table E.3-5.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-37 License Renewal Application E.4 BASELINE RISK MONETIZATION This section explains how Exelon calculated the monetized value of the status quo (i.e., accident consequences without SAMA implementation). Exelon also used this analysis to establish the maximum benefit that could be achieved if all on-line risk were eliminated. E.4.1 OFF-SITE EXPOSURE COST-RISK The baseline annual off-site exposure risk was c onverted to dollars using NRC's conversion factor of $2,000 per person-rem, and discounted to present value using NRC standard formula (NRC 1997a):
W pha =  C x Z pha Where: W pha = monetary value of public health risk after discounting C = [1-exp(-rt f)]/r t f = years remaining until end of facility life = 20 years r = real discount rate (RDR) (as fraction) = 0.03 per year
 
Z pha = monetary value of public health (accident) risk per year before discounting ($ per year) The Level 3 analysis showed an annual off-site population dose risk of 32.61 person-rem. The calculated value for C using 20 years and a 3 percent discount rate is approximately 15.04.
Therefore, 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 (15.04). The calculated off-site exposure cost-risk is estimated to be $980,884. E.4.2 OFF-SITE ECONOMIC COST-RISK The Level 3 analysis showed an annual off-site economic cost-risk of $112,259. Calculated values for off-site economic cost-risks caused by severe accidents over the license renewal period must be discounted to present value as well. This is performed in the same manner as for public health risks and uses the same C value. The resulting value is $1,688,328.
Environmental Report Appendix E  SAMA ANALYSIS Page E-38 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.4.3 ON-SITE EXPOSURE COST-RISK Occupational health was evaluated using NRC met hodology that involves separately evaluating immediate and long-term doses (NRC 1997a). For immediate dose, NRC recommends using the following equation:
Equation 1:
 
W IO = R{(FD IO)S -(FD IO)A} {[1 - exp(-rt f)]/r} Where: WIO = 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 (2.37E-05 events per year)
 
D IO = immediate occupational dose [3,300 person-rem per accident (NRC estimate)]
S = subscript denoting status quo (current conditions)
A = subscript denoting after implementation of proposed action r = RDR (0.03 per year)
 
t f = years remaining until end of facility life (20 years). Assuming F A is zero, the best estimate of the immediate dose cost is:
WIO = R (FD IO)S {[1 - exp(-rt f)]/r}  = 2,0002.37E-05 3,300{[1 - exp(-0.0320)]/0.03}  = $2,352 For long-term dose, NRC recommends using the following equation:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-39 License Renewal Application Equation 2:
W LTO = R{(FD LTO)S -(FD LTO)A} {[1 - exp(-rt f)]/r}{[1 - exp(-rm)]/(rm)}
Where: W LTO = monetary value of accident risk avoided long-term doses, after discounting, $
D LTO = long-term dose [20,000 person-rem per accident (NRC estimate)]  m = years over which long-term doses accrue (as long as 10 years) Using values defined for immediate dose and assuming F A is zero, the best estimate of the cost associated with long-term dose is:
W LTO = R (FD LTO)S {[1 - exp(-rt f)]/r} {[1 - exp(-rm)]/(rm)}  = 2,0002.37E-05 20,000{ [1 - exp(-0.0320)]/0.03} {[1 -exp(-
0.0310)]/(0.0310)}  = $12,318 The total occupational exposure is then calculated by combining Equations 1 and 2 above. The total accident related on-site (occupational) exposure cost-risk (W O) is: W O = W IO + W LTO =  ($2,352 + $12,318) = $14,670 E.4.4 ON-SITE CLEANUP AND DECONTAMINATION COST-RISK The total undiscounted cost of a single event in constant year dollars (C CD) 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. NRC uses the following equation to integrate the net present value over the average number of remaining service years:
PV CD = [C CD/rm][1-exp(-rm)]
Where: PV CD = net present value of a single event Environmental Report Appendix E  SAMA ANALYSIS Page E-40 Three Mile Island Nuclear Station Unit 1 License Renewal Application C CD = total undiscounted cost for a single accident in constant year dollars r = RDR (0.03) m = years required to return site to a pre-accident state The resulting net present value of a single event is $1.3E+09. The NRC uses the following equation to integrate the net present value over the average number of remaining service years:
U CD = [PV CD/r][1-exp(-rt f)] Where: PV CD = net present value of a single event ($1.3E+09) r = RDR (0.03) t f = 20 years (license renewal period) The resulting net present value of cleanup integrated over the license renewal term, $1.95E+10, must be multiplied by the total CDF (2.37E-05) to determine the expected value of cleanup and decontamination costs. The resulting monetary equivalent is $461,912. E.4.5 REPLACEMENT POWER COST-RISK The long-term replacement power cost-risk was determined following NRC methodology in
 
NUREG/BR-0184 (NRC 1997a). The net present value of replacement power for a single
 
event, PV RP, was determined using the following equation:
PVRP = [1.2x10 8 ($-yr)/r]
* [1 - exp(-rt f)]2 Where:  PV RP = net present value of replacement power for a single event, ($) r = RDR (0.03) t f = 20 years (license renewal period)
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-41 License Renewal Application To attain a summation of the single-event costs over the entire license renewal period, the following equation is used:
U RP = [PV RP /r] * [1 - exp(-rt f)]2 Where: U RP = net present value of replacement pow er over life of facility ($-year) After applying a correction factor to account for TMI-1 size relative to the generic reactor described in NUREG/BR-0184 (i.e., 875 megawatt electric/910 megawatt electric) the replacement power costs are determined to be 5.31E+09 ($-year). Multiplying this value by the CDF (2.37E-05) results in a replacement power cost-risk of $125,917. E.4.6 MAXIMUM AVERTED COST-RISK The TMI-1 Maximum Averted Cost-Risk (MACR) is the total averted cost-risk if all internal and external events risk associated with on-line operation were eliminated. This is calculated by summing the following components: Maximum Internal Events Averted Cost-Risk Maximum External Flooding Averted Cost-Risk Maximum External Events Averted Cost-Risk (excluding external flooding)
 
As describ e d in Section E.5.1 , the MACR is used in the SAMA identifica t ion process t o determine the depth of the importance list review. In addition, the MACR is used in the Phase I analysis as a means of screening SAMAs. The following subsections provide a description of how each of these components are calculated and used together to obtain the TMI-1 MACR. E.4.6.1 INTERNAL EVENTS MAXIMUM AVERTED COST-RISK The maximum internal events averted cost-risk is the sum of the contributors calculated in
 
Sections E.4.1 through E.4.5:
Environmental Report Appendix E  SAMA ANALYSIS Page E-42 Three Mile Island Nuclear Station Unit 1 License Renewal Application Maximum Averted Internal Events Cost-Risk Off-site exposure cost-risk
= $980,884 Off-site economic cost-risk = $1,688,328 On-site exposure cost-risk
= $14,670 On-site cleanup cost-risk
= $461,912 Replacement Power cost-risk
= $125,917 Internal Events Maximum Averted Cost-Risk = $3,271,711 This total represents the monetary equivalent of the risk that could be eliminated if all on-line internal events based events could be eliminated for TMI-1. E.4.6.2 EXTERNAL FLOODING EVENTS MAXIMUM AVERTED COST-RISK The same process used to calculate the maximum averted cost-risk for the internal events contributors is used to calculate the maximum averted cost-risk for the external flooding contributors. The external flooding CDF, dose-risk, and economic cost risk estimates are used
 
as input to t he equatio n s presented in Sections E.4.1 through E.4.5. As documented in Section E.2.3.1 , the total external flooding C D F is 8.11E-05 when the contributio n s from all of the flood regimes are summed:
* External floods over 310' msl,
* External floods between 305' msl and 310' msl, and
* External floods below 305' msl The total dose-risk and economic cost-risk for these flood regimes are 177.16 person-rem and
$542,159, respectively, as documented in Secti o n E.3.8. The results of the external flood MACR calculations are provided below: Maximum External Flooding Cost-Risk Off-site exposure cost-risk = $5,328,835 Off-site economic cost-risk = $8,153,861 On-site exposure cost-risk
= $50,177 On-site cleanup cost-risk = $1,579,915 Replacement Power cost-risk
= $430,685 External Flooding Maximum Averted Cost-Risk = $15,543,473 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-43 License Renewal Application E.4.6.3 NON-FLOODING EXTERNAL EVEN TS MAXIMUM AVERTED COST-RISK Finally, the maximum averted cost-risk for external events (excluding external flooding) must be estimated; however, this cost-risk must be estimated based on information in the IPEEE given that current, quantifiable external events models are not available. As d e scribed in S ections E.5.1.5 and E.5.1.6 , these models h ave not been updated to reflect rec e nt plant ch a nges or current PRA techniques. Therefore, the absolute CDF values included in the IPEEE are generally not considered to be directly comparable to the results of the internal events PRA model. The method chosen to account for non-flooding external events in the SAMA analysis is to use a multiplier on the internal events results. In previous SAMA analyses, it has been assumed that the risk posed by external events and internal events is approximately equal. This assumption is not unreasonable unless available analyses indicate that there are external events contributors that present an exceptionally high risk to the site. For TMI-1, external flooding scenarios are considered to present such a risk and are treated separately due to the potentially high frequency of severe flooding events. The relative contributions of the remaining initiators are summarized in the following table:  IPEEE Contributor Summary (No External Flooding) External Event CDF Seismic (LLNL seismic hazard curves) 8.43E-05 Fire 2.16E-05 High Winds 7.77E-07 Aircraft Impact 3.95E-07 Hazardous Chemicals 1.60E-07 Total 1.07E-04 While the CDF total of 1.07E-04 is about a factor of 3 greater than the internal events contribution, a large portion of the CDF is related to seismic risk. The large seismic CDF could be viewed as an indicator that earthquakes, like external floods, may represent an exceptionally high risk t o TMI-1. However, as described in S e ction E.5.1.
6.2.1 , there are several specific issues related to the conservative nature of the seismic analysis that suggest seismic events are not a dominant contributor to the TMI-1 risk profile. As a result, seismic events are grouped with the remaining initiator types.
Environmental Report Appendix E  SAMA ANALYSIS Page E-44 Three Mile Island Nuclear Station Unit 1 License Renewal Application Similarly, the large external events CDF is not considered to be a basis for using a multiplier greater than 2 to account for external events risk due to the high seismic contribution. In fact, the use of unsupported, large multipliers for external events can be detrimental to the SAMA process:
* Over predicting the averted cost-risk of internal events based SAMAs through the use of an inflated multiplier could divert site resources to issues that are not important to the plant,
* Over predicting the averted cost-risk of an external events based SAMA could change the prioritization of addressing cost effective SAMAs away from important issues identified by the internal events model to highly uncertain issues identified by the external events analyses,
* Use of a larger multiplier impacts the MACR, which forces the identification of internal events based SAMAs t h at are not important to plant risk (ref e r to Sections E.5.1.1 and E.5.1.2) and consequen t ially reduces the credibility of the analysis. For these reasons, a multiplier of 2 has been chosen to account for the TMI-1 external events contributions. This implies that the contribution to the MACR from the non-flooding external events is the same as the contribution fr om the internal events model ($3,271,711). E.4.6.4 TMI-1 MAXIMUM AVERTED COST-RISK As stated in Section E.4.6 , the MACR is the total of these th r ee components: Internal Events
= $3,271,711 External Events (excluding External Flooding) = $3,271,711 External Flooding  = $15,543,473 Maximum Averted Cost-Risk = $22,086,895 The MACR is rounded to next highest thousand ($22,087,000) for SAMA calculations. It should be noted that the Phase II cost benefit calculations account for the difference between the rounded MACR and the actual MACR by adding the difference to the averted cost-risk calculated for each SAMA.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-45 License Renewal Application E.5 PHASE I SAMA ANALYSIS The Phase I SAMA ana l ysis, as discussed in S e ction E.1 , i n cludes the d evelopment of the initi a l SAMA list and a coarse screening process. This screening process eliminated those candidates that are not applicable to the plant's design or are too expensive to be cost beneficial even if the risk of on-line operations were completely eliminated. The following subsections provide additional details of the Phase I process. E.5.1 SAMA IDENTIFICATION The initial list of SAMA candidates for TMI-1 was developed from a combination of resources including:
* TMI-1 PRA results
* Industry Phase II SAMAs
* TMI-1 IPE (GPU 1993a)
* TMI-1 IPEEE (GPU  1994) These resources are judged to provide a list of potential plant changes that are most likely to reduce risk in a cost-effective manner for TMI-1. In addition to the "Industry Phase II SAMA" review identified above, an industry based SAMA list was used in a different way to aid in the development of the TMI-1 plant specific SAMA list.
While the industry SAMA review cited above was used to identify SAMAs that might have been
 
overlooked in the development of the TMI-1 SAMA list due to PRA modeling issues, a generic SAMA list was used as an idea source to identify the types of changes that could be used to address the areas of concern identified through the TMI-1 importance list review. For example, if long term DC power availability was determined to be an important issue for TMI-1, the industry list would be reviewed to determine if a plant enhancement had already been conceived that would address TMI-1's needs. If an appropriate SAMA was found to exist, it would be used in the TMI-1 list to address the DC power issue; otherwise, a new SAMA would be developed that would meet the site's needs. This generic list was compiled as part of the development of several industry SAMA analyses and has been provided in Addendum 1 for reference purposes.
Environmental Report Appendix E  SAMA ANALYSIS Page E-46 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.5.1.1 LEVEL 1 TMI-1 IMPORTANCE LIST REVIEW The TMI-1 PRA was used to generate a list of events sorted according to their risk reduction worth (RRW) values. The top events in this list are those events that would provide the greatest reduction in the TMI-1 CDF if the failure probability were set to zero. The events were reviewed down to the 1.010 level, which was chosen because it corresponds to the definition of a risk significant event, as defined in the PSA Applications Guide. [EPRI 1995] An alternate method of establishing the lower review threshold would be to correlate the minimum expected SAMA implementation cost to an RRW value. For TMI-1, the minimum expected cost of implementation is believed to be a procedure change. The cost of procedure changes can vary depending on the type of procedure being modified and the scope of the changes, but a representative value is considered to be about $50,000, which is supported by previous industry cost estimates for procedure modifications [CPL 2004]. For TMI-1, the RRW value corresponding to $50,000 is about 1.008 (excluding External Flooding contributions). This can be demonstrated by reducing the CDF, dose-risk and off-site economic cost-risk by a factor of 1.008, which corresponds to an event with Level 1 and Level 2 based RRW values of just under 1.008. The corresponding internal events based averted cost-risk would be $25,966. Applying a factor of 2 to estimate the potential impact of external events (refer to Section E.4.6) results in a cost-risk of
$51,932. This is approximately equ a l to the assumed minimum expected cost of implementation. While the RRW value of 1.008 is not exactly equal to the 1.010 established by the PSA Applications Guide definition of risk significance, the RRW threshold values are consistent and the use of 1.010 is considered to be adequate for this analysis. The External Flooding contributions are excluded from the calculations establishing the RRW review threshold because the identification and quantification processes for External Flooding
 
SAMAs are performed separate from the internal events model.
Table E.5-1 documents the dispositi o n of each event in the L e vel 1 TMI-1 RRW list with RRW values of 1.010 or greater. Note that the review of each event involves a detailed evaluation of the cutsets including the event to identify the factors that make the event important.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-47 License Renewal Application E.5.1.2 LEVEL 2 TMI-1 IMPORTANCE LIST REVIEW A similar review was performed on the importance listings from the Level 2 results. In this case, a composite importance file based on all release categories except RC9 was used to identify potential SAMAs. This method was chosen to prevent high frequency-low consequence events from dominating the importance listing. While RC9 contributes about 13 percent of the dose-risk, that small contribution depends on over 66 percent of the Level 2 frequency, which would heavily bias the importance list toward RC9 contributors. The Level 2 RRW values were also reviewed down to the 1.010 level. As described for the Level 1 RRW list, events below the 1.010 threshold value are not "risk significant" and are not expected to yield cost beneficial SAMAs.
Table E.5-2 documents the dispositi o n of each event in the L e vel 2 TMI-1 RRW list with RRW values greater than 1.010. E.5.1.3 INDUSTRY SAMA ANALYSIS REVIEW The SAMA identification process for TMI-1 is primarily based on the PRA importance listings, the IPE, and the IPEEE. In addition to these plant-specific sources, selected industry SAMA submittals were reviewed to identify any Phase II SAMAs that were determined to be potentially cost beneficial at other plants. These SAMAs were further analyzed and included in the TMI-1 SAMA list if they were considered to address potential risks not identified by the TMI-1
 
importance list review. While many of the industry SAMAs reviewed are ultimately shown not to be cost beneficial, some are close contenders and a small number have been estimated to be cost beneficial at other plants. Use of the TMI-1 importance ranking should identify the types of changes that would most likely be cost beneficial for TMI-1, but review of selected industry Phase II SAMAs may capture potentially important changes not identified for TMI-1 due to PRA modeling differences or SAMAs that represent alternate methods of addressing risk. Given this potential, it was considered prudent to include a review of selected industry Phase II SAMAs in the TMI-1 SAMA identification process. Phase II SAMAs from the following U.S. nuclear power sites have been reviewed:
* Turkey Point Environmental Report Appendix E  SAMA ANALYSIS Page E-48 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Arkansas Nuclear One, Unit 1
* Palisades
* D.C. Cook, Units 1 and 2
* Susquehanna Units 1 and 2
* Fitzpatrick Four PWR and two boiling water reactor (BWR) sites were chosen from available documentation to serve as the Phase II SAMA sources. Few of the Phase II SAMAs from these sources were included in the initial TMI-1 SAMA list. Many of the industry Phase II SAMAs were already represented by other SAMAs in the TMI-1 list, were known not to impact important plant systems, or were judged not to have the potential to be close contenders for TMI-1. These SAMAs were not considered further. The following provides a summary of some of the issues considered during the review of the industry SAMAs. E.5.1.3.1 Turkey Point Turkey Point used a generic SAMA list as its starting point and few plant specific insights were available that might pertain specifically to B&W PWRs. In addition, only limited averted cost information was provided for the SAMAs and no changes were identified as cost beneficial, which made review of the list difficult. One SAMA had the potential to address a portion of TMI-1 risk in an inexpensive manner, but equipment limitations precluded its direct application to TMI-1:
* Turkey Point SAMA 111 - This SAMA suggests using Firewater as an alternate means of providing makeup to the steam generators. The prominent Level 1 cases involving loss of SG makeup flow at TMI-1 are SBO cases where the seals are in jeopardy. Providing alternate secondary side makeup without addressing the seal LOCA would not have a large impact on risk for TMI-1. In order for the use of Firewater to address important TMI-1 sequences, it would have to be capable of providing SG makeup early in the accident sequence and be combined with the installation of high temperature, damage resistant seals so that primary side inventory is not lost. Given that early SG makeup would require a pressure greater than the 130 psig available fr om the Fire Service Water system, this SAMA is not considered to be practical for TMI-1 and it is not included on the SAMA list. For Level Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-49 License Renewal Application 2, a large contributor to dose-risk is the failure to maintain water in the SGs to provide fission product scrubbing and for preventing induced tube rupture events. These events are considered to be best addressed by the addition of an independent auxiliary feedwater system, which is included as SAMA 22 based on the TMI-1 Level 2 importance list review. E.5.1.3.2  ANO-1 While a generic SAMA list similar to the one used for Turkey Point was used in the ANO-1 SAMA submittal, one SAMA was found to be cost beneficial for ANO-1. This SAMA addresses the operator action to swap to recirculation mode, which was identified as an important contributor to TMI-1 risk:
* ANO-1 SAMA 129 suggests emphasizing a timely swap to recirculation mode in operator training and procedures. Theoretically, more emphasis could be placed on this well recognized issue for TMI-1, but in order to achieve a meaningful risk reduction based on training improvements, a specific deficiency would have to be identified in the TMI-1 training materials or procedures that could be rectified. No such deficiency has been identified based on the information available in the HRA. A SAMA has been proposed for TMI-1 to automate the swap to recirculation (SAMA 15), which would remove the operator from the primary role in the action. This is considered to be a more effective means of reducing the risk related to recirculation initiation failures for TMI-1. No additional SAMA related to improved training for swap to recirculation mode has been added to the TMI-1 SAMA list. E.5.1.3.3  Palisades Palisades identified several cost beneficial SAMAs; however, most of the changes were related to plant specific issues that are not applicable to TMI-1. Potential exceptions include adding the capability to operate EFW without power support and installation of a diesel motor to drive an EFW pump. These types of changes were shown to have a large impact on risk for Palisades and subsequent review of the plant design yielded the conclusion that the most effective means of addressing LOOP/SBO risk for the site was the installation of an additional EDG. For TMI-1 these three issues are dispositioned as follows:
* SAMA 2 addresses the use of a portable generator to allow extended EFW operation. It is combined with RCP seal upgrades as the important contributors including prolonged EFW operation are those in which seal integrity is challenged. This is considered to be the most appropriate means of addressing prolonged EFW operation for TMI-1 and no additional Environmental Report Appendix E  SAMA ANALYSIS Page E-50 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMAs are suggested.
* Installation of a diesel engine to drive an EFW pump would improve the capability of TMI-1 to address SBO cases in which EFW has failed. Other industry investigations of this SAMA have concluded that connecting a diesel motor to an EFW pump would be easier/cheaper for a turbine driven pump than for a motor driven pump; however, for TMI-1, the initial TD EFW failure may preclude the use of the pump even with the diesel engine. For improved effectiveness in the important TMI-1 scenarios, the diesel engine should be connected to a motor driven EFW pump or a unique diesel driven pump should be used. In addition, this type of change needs to be accompanied by the installation of the high temperature, damage resistant seals to preclude the seal LOCA that will result from an SBO. Without securing primary side integrity, extended secondary side cooling would provide limited benefit. Finally, a portable generator would be required to power SG level instrumentation for effective level control. While TMI-1 SAMA 10 already addresses SBO cases with EFW failures, this diesel driven pump option provides an alternate approach to the issue and it has been included on the SAMA list for evaluation (SAMA 24).
* The addition of an EDG at Palisades as a result of the SAMA analysis would bring the total number of EDGs at the plant to three, which is equivalent to the current TMI-1 configuration.
Some benefit could be gained through the installation of a fourth EDG for TMI-1, but common cause failures would limit the benefit and there are more cost effective changes that could be made to the existing EDG configuration that would greatly reduce risk (i.e.,
SAMA 1). Even with the inclusion of SAMAs 11 and 24 already on the SAMA list, a SAMA suggesting the addition of another EDG has been added to the TMI-1 SAMA list as an alternate means of reducing SBO risk (SAMA 25). E.5.1.3.4 D.C. Cook The D.C. Cook SAMA analysis showed that 5 different types of changes were determined to be cost beneficial. In three of the five areas, multiple SAMAs are identified as potentially cost beneficial and no single approach is identified as the most appropriate for D.C Cook. These risk areas were reviewed for TMI-1 and it was determined that the issues are already adequately addressed by the TMI-1 SAMA list or that the risk areas were not important contributors for the
 
site:
* Minimize Consequences of RCP seal LOCAs:  The TMI-1 SAMA list includes multiple Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-51 License Renewal Application SAMAs addressing seal LOCA prevention, including the use of new seals (SAMA 2) and a means of providing an alternate heat sink for the thermal barrier cooling system (SAMA 7).
* Minimize Consequences of Loss of HVAC:  TMI-1 does not require HVAC for successful operation of the plant during the 24 hour mission time considered in the PRA.
* Remove Dependence of Distributed Ignition System on AC Power:  TMI-1 does not have igniters in the containment. A battery backed hydrogen ignition system could be added, which is included as SAMA 19 based on the TMI-1 Level 2 importance list review.
* Minimize Consequences of AC Bus Failures:  AC cross-ties are proposed in the D.C. Cook SAMA analysis as a means of reducing the contribution of bus failures. It is not clear how a cross-tie would mitigate the bus failure cited in the analysis, but for TMI-1 bus failures are not large contributors to risk. The availability of the SBO EDG and its capability to be aligned to either division reduces the risk of these events.
* Improve Recovery from ISLOCA:  For TMI-1, ISLOCA is dominated by DHR suction path failures after leak or rupture of valves DH-V-1 and DH-V-2. While the TMI-1 ISLOCA analysis does not take credit for any potentially mitigating actions, no actions that could reliably terminate the event are believed to be available. For example, 1) the isolation of DH-V-3 may not isolate the break or additional breaks may occur after isolation, 2) reduction of primary system pressure may reduce the flow out of the break, but it would not stop it, and 3) refill of the BWST does not place the plant in a stable state and the impacts of aux building flooding would have to be addressed. A SAMA was added to the TMI-1 list to extend the high pressure boundary in the DHR suction lines to include an additional isolation valve based on the TMI-1 Level 2 importance list review (SAMA 20). E.5.1.3.5 Susquehanna The Susquehanna SAMA analysis showed that five SAMAs were potentially cost beneficial when considered independently. When considerati on was given to the overlapping benefits of the SAMAs and limits of the assessment process, only two were considered to be likely candidates for implementation. For TMI-1, it was determined that the issues are already adequately addressed by the TMI-1 SAMA list or that the risk areas were not important contributors for the site:
* SSES SAMAs 2a and 2b (4kV AC Cross-ties):  The availability of the SBO EDG, which can Environmental Report Appendix E  SAMA ANALYSIS Page E-52 Three Mile Island Nuclear Station Unit 1 License Renewal Application be aligned to either division, serves a purpose similar to that of an AC cross-tie and minimizes the benefit of any cross-tie SAMAs. The existing cross-tie capability is not credited in the model.
* SSES SAMAs 5 and 6 (Additional/Auto Aligning Portable 480V AC Generators): The use of a portable 480V generator is suggested in TMI-1 SAMA 2 in combination with the installation of improved RCP seals. This change is considered to be the most appropriate for the TMI-1 design.
* SSES SAMA 3 (Staggered Depressurization):  This is a 2 unit BWR issue that is not applicable to TMI-1. No additional SAMAs have been added to the TMI-1 SAMA list based on a review of these
 
SAMAs. E.5.1.3.6 Fitzpatrick The Fitzpatrick SAMA analysis identified two types of SAMAs as potentially cost beneficial. The SAMAs related to extending DC power availability are addressed by TMI-1 SAMA 2 and the SAMA related to providing alternate EDG HVAC is not applicable to TMI-1 as HVAC is not required for the 24 hour PRA mission time. No additional SAMAs have been added to the TMI-1 SAMA list based on a review of these SAMAs. E.5.1.3.7 Industry SAMA identification Summary The important issues for TMI-1 are considered to be addressed by the SAMAs developed through the PRA importance list review. Further, the plant changes suggested as part of that review were developed to meet the specific needs of the plant such that those SAMAs are more likely to provide effective means of risk reduction than SAMAs taken from other sites. However, effort was made to review other industry SAMA analyses to determine if other sites identified plant changes that could be cost beneficial for TMI-1. While it was found that other plants had developed SAMAs that addressed areas of conc ern for TMI-1, only two have been identified that could be adapted for inclusion in the TMI-1 SAMA list. While these SAMAs can be considered unique, the SAMAs only propose alternate means of addressing issues already
 
targeted by other TMI-1 SAMAs:
* Install Damage Resistant, High Temperature RCP Seals with a Diesel Engine as an Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-53 License Renewal Application Alternate Drive for an EFW Pump and Portable Generator for Level Control Instrumentation (SAMA 24).
* Install an Additional EDG (SAMA 25). E.5.1.4 TMI-1 IPE The TMI-1 IPE generated a list of risk-based insights and potential plant improvements.
Typically, changes identified in the IPE process are implemented and closed out; however, there are some items that are not completed within the industry due to high projected costs or other criteria. Because the criteria for implementation of a SAMA may be different than what was used in the post-IPE decision-making process, these recommended improvements are re-examined in this analysis. As a result of the IPE, five potential plant improvements were identified and considered for implementation at the plant. The following table summarizes the status of these plant improvements.
Environmental Report Appendix E  SAMA ANALYSIS Page E-54 Three Mile Island Nuclear Station Unit 1 License Renewal Application Description of Potential Enhancement Status of Implementation Disposition Provide additional procedural guidance to direct operators to throttle low pressure injection prior to swapping the pump suction source from the BWST to the containment sump. Ensuring this step is taken will reduce the likelihood of incurring pump damage during the transition. Implemented. Current procedures direct throttling of LPI flow after
 
injection initiation as well as actions to mitigate pump cavitation in the event that the initial throttling steps do not preclude cavitation. No further review required. Enhance accident management guidelines for SGTR events to direct isolation of the failed OTSG and cooldown of the primary system using the intact OTSG. This is considered an effective means of mitigating SGTR scenarios. Implemented. B&W Generic Emergency Operating Guidelines direct OTSG isolation on a number of signals, including high radiation and SG level, which are indicators of SGTR events. Cooldown of the reactor is also part of the generic guidance; therefore, the intent of this SAMA is met by the existing procedures. No further review required. For those SGTR cases in which isolation of the ruptured SG is not possible, inventory loss may continue through the ruptured OTSG. Updating the accident management guidelines to direct refill of the BWST to keep pace with the RCS inventory loss would help mitigate the evolution until other steps to stabilize the plant could be
 
taken. Implemented No further review required. Update the accident management guidelines to direct the operators to verify closure of the MU-14 valves after the transition to "piggyback recirculation mode" from high pressure injection mode. This would provide additional assurance that pathways to the BWST and the environment are isolated when this mode of recirculation is used. Implemented No further review required.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-55 License Renewal Application Description of Potential Enhancement Status of Implementation Disposition Consider including the following operator actions in the Licensed Operator Requalification training
 
Program: 1. Switchover to reactor sump recirculation following a
 
LOCA 2. Refilling the BWST given SGTR 3. Properly throttling HPI flow after ES actuation 4. Holding open or reopening RCP seal injection valve MU-V-20 on loss of instrument air 5. Tripping RCPs before seal damage after loss of NSCCW 6. Taking actions to prevent boron concentration when in recirculation following a LOCA Partial implementation:
: 1. The action to swap to recirculation following a LOCA is included in requalification training, most recently in year 2005.2. No specific training has been identified for BWST
 
refill in an SGTR. 3. The action to throttle HPI flow to prevent overcooling/overpressurization is included in requalification training, most recently in year 2006.4. No specific training has been identified related to re-opening MU-V-20 on loss of IA. 5. The action to trip RCPs before seal damage on loss of NSCCW is included in requalification training, most recently in years 2005 and 2006. 6. The action to prevent boron concentration effects while in recirc mode after a LOCA is included in requalification training, most recently in year 2005.The actions suggested for inclusion in the TMI-1 training program were based on the importance of the actions as evaluated in the IPE model. As the PRA is a living analysis, there is a potential for the importance of the operator actions to change based on the use of new failure data, inclusion of logic to reflect plant changes, application of improved modeling practices that remove conservatism, or elimination of errors. The importance list review performed for the SAMA analysis will identify the most important actions modeled in the current TMI-1 PRA. While no requalification training appears to be performed for items 2 or 4 from the list of actions suggested for inclusion in the requalification training by the IPE, the current PRA model does include these events:
* BWST-HRE27-HTKOA: FAILURE TO REFILL BWST (SPLIT FRAC REV) (HRE27
 
in the IPE)
* INHINJ4_MUHHVCOA:
OPERATOR REOPENS MU-V20 (HINJ4 in the IPE)
As a result, the SAMA process will address these actions, if necessary, and inclusion of a SAMA to add these actions to the requalification program independently from the importance list review is not required.
All of the plant changes proposed by the IPE have either been implemented or are addressed by the SAMA process. No SAMAs are included on the TMI-1 SAMA list to address IPE insights. E.5.1.5 TMI-1 IPEEE Similar to the IPE, any insights that were previously dispositioned based on non-SAMA criteria are re-examined as part of this analysis. In addition, any insights that are in the process of Environmental Report Appendix E  SAMA ANALYSIS Page E-56 Three Mile Island Nuclear Station Unit 1 License Renewal Application being addressed are examined as their resolutions could be important to the disposition of some SAMAs. The IPEEE was used to identify these items. The following table summarizes the status of the potential plant enhancements resulting from the IPEEE processes and their treatment in the SAMA analysis. As can be seen, several unimplemented insights have been identified and included on the SAMA list: Description of Potential Enhancement Status of Implementation Disposition Install a flood safe means of providing 480V AC power and pumps to provide RCP seal cooling and makeup to the steam generators. Implemented While implemented, the design has been reviewed to determine if additional changes could be made to
 
improve reliability. See S e ction E.5.1.6.4. Load centers 1P, 1R, 1S, and 1T: add gusset weld reinforcements to improve seismic ruggedness.
Not imple m ented I nclu d ed a s SAMA 27. S e e s e ction E.5.1.6.2.2.
Install additional supports for the main control room ceiling to prevent failure in seismic events.Implemented No further review required. Install a restraint on penetration pressurization tank PP-T-1A to prevent seismic interaction with reactor building purge inlet isolation valve AH-V-1D. Implemented No further review required. Modify the diesel fire pump battery and fuel oil tank supports to increase their seismic ruggedness.
Not imple m ented I nclu d ed a s SAMA 30. See secti o n E.5.1.6.2.2. Modify the anchorage for the decay heat service heat exchangers (DC-C-2A(B)) to improve their seismic ruggedness.
Not imple m ented I nclu d ed a s SAMA 28. See secti o n E.5.1.6.2.2. Modify the anchorage for the EDG air receivers to improve their seismic ruggedness. Implemented No further review required. An effort was also made to use the IPEEE to develop new SAMAs based on a review of the original results. However, the TMI-1 IPEEE was not maintained as a "living" analysis. This limits the capability of the models that make up the IPEEE as they do not include the latest PRA practices nor do they necessarily represent the current plant configuration or operating characteristics. The fact that the models are not currently in a quantifiable state presents further difficulty because the results are limited to what has been retained from the original analysis.
These factors limit the qualitative insights and quantitative estimates that can be made with regard to external events contributors. On a larger scale, given that the industry has generally not pursued external events modeling at a level consistent with internal events models, the technology for external events analysis is not as robust or refined. The result is that the CDF Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-57 License Renewal Application values yielded by the internal and external ev ents models are not necessarily comparable. External events models are considered to be useful tools for identifying important accident sequences and mitigating equipment, but the quantitative results should not be directly combined with those from the internal events models. In this analysis, external events contributions are estimated using a multiplier on the internal events results for the reasons described above. The exception is the treatment of external flooding. Finally, it was necessary to review the changes to the site and surrounding area that were implemented after the completion of the IPEEE to determine if the changes could impact the conclusions of those analyses. The only changes identified with the potential to impact the conclusions of the IPEEE are the installation of the security towers and security fencing on the site grounds. In high wind events,
* Security towers may be sources for wind generated missiles, and
* Security fencing could blow into areas where they may prevent access to equipment required for mitigating actions. The security towers are considered to be unlikely sources for wind generated missiles due to the fact that their design requires them to be able to withstand vehicle impact. With respect to the security fence issue, the only potentially important action identified that normally requires travel in areas where the fences could be an issue is the start of the SBO EDG. However, there is an access door to the SBO EDG building in the Unit 2 structure that could be used, if
 
required. Finally, as described in S e ction E.5.1.
6.3 , the max i mum a v ert e d cost risk f or high wind related scenarios is well under the minimum expected cost of implementation of $50,000. This indicates that SAMAs that only impact high wind risk can not be cost beneficial. As a result, it has been concluded that plant changes subsequent to the completion of the IPEEE do not invalidate the docketed results. E.5.1.6 USE OF EXTERNAL EVENTS IN THE TMI-1 SAMA IDENTIFICATION PROCESS The IPEEE was used in the TMI-1 SAMA analysis prim arily to identify the highest risk accident sequences and the potential means of reducing the risk posed by those sequences. The types of events considered in the TMI-1 external events analysis were identified by Supplement 4 of
 
Generic Letter 88-20 (NRC 1991) and included:
Environmental Report Appendix E  SAMA ANALYSIS Page E-58 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Internal Fir e s (Section E.5.1.6.1)
* Seismic Ev e nts (Section E.5.1.6.2)
* High Wind Events (Section E.5.1.6.3)
* External Flooding (Section E.5.1.6.4)
* Transportat i on and Nearby Facility Accidents (Section E.5.1.
6.5) Based on the TMI-1 review, no additional hazards were identified for analysis in the IPEEE. The type of information available for the initiators that were evaluated by TMI-1 varied due to the manner in which they were addressed in the IPEEE. For instance, the fire analysis used an approach that combined the deterministic evaluation techniques from the EPRI Fire Induced Vulnerability Evaluation (FIVE) methodology with classical PRA techniques. The TMI-1 seismic analysis was performed using modified versions of the TMI IPE model to address seismic impacts on the plant's accident response capabilities. Core damage frequencies were also estimated for external flooding, high wind events, and transportation and nearby facility accidents. Due to limitations of the modeling processes, however, the results of these kinds of analyses are not necessarily compatible with those of the internal events analysis. As a result, each of the external event contributors must be considered in a manner suiting the type of analysis performed. A summary of the review process used to identify SAMAs is provided for each of the external event types listed above. E.5.1.6.1 Internal Fires As discussed above, the techniques used to model external events vary according to the type of initiator being analyzed. The TMI-1 Fire Model shares many of the same characteristics as the internal events model, but limitations on the state of technology produce results that are typically more conservative than the internal events model. The following summarizes the fire PRA topics where quantification of the CDF may introduce different levels of modeling uncertainty than the internal events PRA. In general, fire PRAs are useful tools to identify design or procedural items that could be clear areas of focus for improving the safety of the plant. Fire PRAs use a structure and quantification technique similar to that used in the internal events PRA. Since less attention Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-59 License Renewal Application historically has been paid to fire PRAs, conserva tive modeling is common in a number of areas of the fire analysis to provide a "bounding" methodology for fires. This concept is contrary to the base internal events PRA, which has had more analytical development and is judged to be closer to a realistic assessment (i.e., best estimate) of the plant. There are a number of fire PRA topics involving technical inputs, data, and modeling that prevent the effective comparison of the CDF between the internal events PRA and the fire PRA. These areas are identified as
 
follows: PRA Topic Comment Initiating Events: The frequency of fires and their severity are generally conservatively overestimated. A revised NRC fire events database indicates the trend toward lower frequency and less severe fires. This trend reflects the improved housekeeping, reduction in transient fire hazards, and other improved fire protection (FP) steps at plants. System Response: FP measures such as sprinklers, CO 2, and fire brigades may be given minimal (conservative) credit in their abilit y to limit the spread of a fire. Sequences: Sequences may subsume a number of fire scenarios to reduce the analytic burden. The subsuming of initiators and sequences is done to envelope those sequences included. This results in additional conservatism. Fire Modeling: Fire damage and fire spread are conservatively characterized. Fire modeling presents bounding approaches regarding the immediate effects of a fire (e.g., all cables in a tray are always failed for a cable tray fire) and fire propagation.
HRA: There is little industry experience with crew actions under conditions of the types of fires modeled in fire PRAs. This has led to conservative characterization of crew actions in fire PRAs. Because the CDF is strongly correlated with crew actions, this conservatism has a profound effect on the calculated fire PRA results. Level of Detail: The fire PRAs may have reduced level of detail in the mitigation of the initiating event and consequential system damage. Quality of Model: The peer review process for fire PRAs is not as developed as internal events PRAs. For example, no industry standard, such as NEI 00-02, existed for the structured peer review of a fire PRA. This may result in less assurance of the realism of the model. In addition to modeling limitations, the fire PRA may be subject to more modeling uncertainty than the internal events PRA evaluations. While the fire PRA is generally self-consistent within its calculational framework, the fire PRA does not compare well with internal events PRAs because of the number of conservative assumptions that have been included in the fire PRA process. Therefore, the use of the fire PRA results as a reflection of CDF may be inappropriate. Any use of fire PRA results and insights should consider areas where the "state of the art" in fire
 
PRAs is less evolved than other PRA topics.
Environmental Report Appendix E  SAMA ANALYSIS Page E-60 Three Mile Island Nuclear Station Unit 1 License Renewal Application While the ability to directly compare the results of the internal events and fire models is limited, information is available that may be used to identify the most important contributors for TMI-1. The IPEEE provides some information related to equipment failures by fire scenario. This information has been summarized in the table below for the five fire scenarios that were not screened on low CDF.
Fire Area/
Scenario Description CDF Major Equipment Failed CB-FA-2d East Inverter Room 4.94E-06/yr Vital instrument bus ATA, battery chargers 1A and 1C, inverters 1A, 1C, and 1E, and control cables for 4.16kV AC emergency bus 1D. CB-FA-2e West Inverter Room 5.81E-06/yr Battery chargers 1B and 1D, inverters 1B and 1D, and control cables for 4.16kV AC emergency bus 1E. CB-FA-3a 1D Switchgear Room 3.94E-06/yr 4.16kV AC emergency bus 1D. CB-FA-3b 1E Switchgear Room 4.96E-06/yr 4.16kV AC emergency bus 1E. CB-FA-4b Control Room - Console CR 1.96E-06/yr
* RCS inventory control and injection: makeup pumps MU-P-1B, MU-P-1C, and
 
injection valves  MU-V-16C, D
* Nuclear Service River Water Pumps NR-P-1B, C
* Nuclear Services Closed Cycle Cooling Water Pumps NS-P-1B, C
* RCP seal injection and cooling: ICCW pumps, NR-V-10A, and B, NR-V-15A and
 
B
* Train B of DHR, DR, and DCCW, including DH-V-4B, 5B, 6B, and 7B.
* ESAS manual actuation for train B
* Operation of Containment spray and fans
* Essential AC power: EDG 1A controls, EDG 1B controls, SBO DG controls, 1D 4.16kV AC bus controls, 1E 4.16kV AC bus controls. Since the fire IPEEE is based on a progressive screening methodology, the CDF values for the fire areas presented above should not be arbitrarily added. Due to the differing levels of detail required to screen the various areas from further consideration, there can be significant conservative assumptions implicit in some of the final values, whereas some of these conservative assumptions may have been relaxed for more detailed analysis. Given this perspective, the CDF for these fire areas could be estimated as 2.16E-05/yr. The table above Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-61 License Renewal Application demonstrates that the CDF is distributed more or less evenly among the non-screened fire scenarios and that there are no dominant scenarios that contribute nearly all of the fire risk. In addition, while fires in each of these areas may impact a wide range of equipment, damage is typically limited to a single division. As a result, redundant equipment is often available to mitigate the fire events. Further discussion is provided for each of the fire area/scenarios below. E.5.1.6.1.1 CB-FA-2d: East Inverter Room Fires in the East Inverter room essentially fail the "A" division of AC and DC power. Random failures of specific "B" train equipment in conjunction with the fire event result core damage.
Providing a means of maintaining primary side integrity and secondary side cooling without electric support is a potential means of reducing the risk of these fire scenarios. Given that two portable 480V AC generators are already available at TMI-1, one to support the severe flooding guidelines and one for general plant use, the TMI-1 turbine driven EFW pump would be capable of providing secondary side cooling for extended periods without 4.16kV AC power if one of the 480V AC generators was used to power one of the 125V DC battery chargers (for level instrumentation/valve and pump control). Installation of the forthcoming Westinghouse type high temperature, damage resistant seals would virtually prevent seal LOCAs and maintain primary side inventory for extended periods (SAMA 2). Providing power to a 125V DC battery charger is considered to be required because the 125V DC system supports the vital 120V AC power supply for the OTSG level indicators. No credit is taken for operation of the TD EFW pump without level indication. Some of the risk from fires in this room was identified in the IPEEE as resulting from damage to cables that run over ignition sources. Early insights from the work being performed for the TMI-1 fire model update indicate that there are no cables over ignition sources in this area that would be problematic. As a result, no SAMA is suggested to re-route or wrap the cables in this area; however, the core damage frequency for this room is conservatively not reduced to reflect this insight. E.5.1.6.1.2 CB-FA-2e:
West Inverter Room Fires in CB-FA-2e are similar to fires in CB-FA-2d. A fire in the West Inverter room essentially fails the "B" division of AC and DC power. Random failures of the "A" train equipment typically result in loss of the corresponding systems and core damage will ensue. Providing a means of Environmental Report Appendix E  SAMA ANALYSIS Page E-62 Three Mile Island Nuclear Station Unit 1 License Renewal Application maintaining primary side integrity and secondary side cooling without electric support is a potential means of reducing the risk of these fire scenarios. Given that two portable 480V AC generators are already available at TMI-1, one to support the severe flooding guidelines and one for general plant use, the TMI-1 turbine driven EFW pump would be capable of providing secondary side cooling for extended periods without 4.16kV AC power if one of the 480V AC generators was used to power one of the 125V DC battery chargers (for level instrumentation/valve and pump control). Installation of the forthcoming Westinghouse type high temperature, damage resistant seals would virtually prevent seal LOCAs and maintain primary side inventory for extended periods (SAMA 2). Providing power to a 125V DC battery charger is considered to be required because the 125V DC system supports the vital 120V AC power supply for the OTSG level indicators. No credit is taken for operation of the TD EFW pump without level indication. Some of the risk from fires in this room is from damage to cables that run over ignition sources. If the cable trays were re-routed away from the electrical equipment that they currently overpass or if the cables were wrapped with fireproof material, the consequences of fires in the inverter room equipment could be reduced (SAMA 26). E.5.1.6.1.3 CB-FA-3a a nd CB-FA-3b: 1D and 1E Switchgear Rooms The only critical equipment located in these areas is the switchgear itself. Due to the layout of the switchgear, with main distribution buswork running through each major cubicle, a fire in virtually any cubicle could short the main buses to ground, disabling the entire train of switchgear. Even if the main buses are not failed, the fire brigade may require the bus to be de-energized to allow fire suppression. It may theoretically be possible to improve the response of the fire brigade or provide some automated fire mitigation system to prevent the spread of the initiating fire; however, the fire would cause some damage to the switchgear before the mitigating actions could be initiated and the extinguishing method itself could cause additional damage to the switchgear. Due to the uncertainty related to potential switchgear damage, mitigating the effects of a fire in this area is considered to be a more appropriate means of addressing the fire risk than attempting to mitigate the fire itself. Given that two portable 480V AC generators are already available at TMI-1, one to support the severe flooding guidelines and one for general plant use, the TMI-1 turbine driven EFW pump would be capable of providing secondary side cooling for extended Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-63 License Renewal Application periods without 4.16kV AC power if one of the 480V AC generators was used to power one of the 125V DC battery chargers (for level instrumentation/valve and pump control). Installation of the forthcoming Westinghouse type high temperature, damage resistant seals would virtually prevent seal LOCAs and maintain primary side inventory for extended periods (SAMA 2).
Providing power to a 125V DC battery charger is considered to be required because the 125V DC system supports the vital 120V AC power supply for the OTSG level indicators. No credit is taken for operation of the TD EFW pump without level indication. E.5.1.6.1.4 CB-FA-4b: Control Room, Console CR A main control room fire in console CR results in the loss of a variety of equipment and will likely force evacuation of the area to the remote shutdown panel (RSP). The RSP contains only a subset of the controls found in the main control room that were determined to be required to control the plant assuming that all of the equipment on the panel is available. In the case of a Console CR fire, some of this critical equipment is considered to be failed as a result of the fire, including NSRW pump NR-P-1C, NSCCW pump NS-P-1C, and train "B" of DHR/DR.
Consequently, the RSP does not provide an adequate means of controlling the reactor in these scenarios. A potential means of addressing this issue would be to expand the RSP to include both trains of the safe shutdown equipment. However, this option creates an area where a single fire could disable both trains of safety equipment. For this reason, enhancing the RSP in this way is not
 
suggested. Other SAMAs could be developed to address risk in this area, but given that the main control room is always manned and that no credit was taken for manual detection of a fire, the contribution from this fire area is considered to be overestimated and no SAMAs are believed to be required for main control room fires. Even if main control room fires could only be detected and extinguished 90 percent of the time, taking this credit in the IPEEE would have reduced the contribution of main control room fires to 1.96E-7, which would have been below the screening criteria for retention. E.5.1.6.1.5 Fire SAMA Identification Summary Based on the review of the TMI-1 fire area results, two SAMAs have been identified as potentially cost beneficial methods of reducing fire risk:
Environmental Report Appendix E  SAMA ANALYSIS Page E-64 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for Extended EFW Operation (SAMA 2),
* Re-route Cables in Inverter Rooms (SAMA 26), Any SAMAs that improve the plant response to an accident have the potential for reducing fire risk through the same mechanisms; however, these SAMAs are also considered to explicitly address the fire scenarios identified above. While SAMA 2 has been identified as potential means of reducing fire risk, it was also identified based on the internal events importance list and is not unique to the fire review. E.5.1.6.2 Seismic Events In response to Generic Letter 88-20, Supplement 4 (NRC 1991), TMI-1 prepared a seismic PRA (SPRA) to asses seismic risk at the site. The SPRA considered site specific seismic event frequencies in conjunction with the plant specific response to quantify a CDF using a modified version of the IPE risk model. The baseline case was developed using seismic event frequencies developed by EPRI (EPRI 1989), but also quantified risk based on the frequencies estimated by Lawrence Livermore National Labs (NRC 1994). The results from the Lawrence Livermore National Lab (LLNL) sensitivity are assumed to be the baseline results for the purposes of the SAMA analysis.
E.5.1.6.2.1 Seismic Modeling Overview As with the Fire model, the TMI-1 seismic model was not maintained as a living model. As a result, the state of knowledge, use of current PRA techniques, and subsequent plant changes are not reflected in the SPRA results. However, the development of a full SPRA likely provided a more thorough evaluation of seismic risk than a seismic margins analysis, which many plants in the industry used for the IPEEE. The following steps summarize the seismic modeling process used for TMI-1:
: 1. Determination of site specific seismicity characteristic. This step involves the development of the frequencies of occurrence and magnitude of seismic events for the TMI site. Site structure analysis is also performed. The resulting frequencies and magnitudes of seismic events are the initiating events for the SPRA. Site structure responses are input into Step 5 where the capacities of the components which impact risk are calculated.
: 5. Identification of those components important to plant safety , including equipment, structures and procedures. The Level 1 PRA developed for TMI-1 is utilized to Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-65 License Renewal Application determine those components which impact risk. Other studies such as the TMI Environmental and External Hazards Report and the USI A-46 Safe Shutdown Equipment List are used to ensure that the list of components which impact risk is
 
comprehensive.
: 6. An initial plant walkdown of the identified systems and components is performed. Any plant seismic system interactions and unique plant features which may impact risk are identified.
: 7. Develop plant logic models. The plant logic models are developed using the Level 1 TMI-1 PRA with the addition of the failure rates (fragilities) of components due to seismically initiated events. A "pre-tree" approach is utilized to ensure that independent as well as seismic failures are accounted for in the logic model.
: 8. A second plant walkdown is performed to verify plant seismic response models and to collect data to determine component capacities.
: 9. Analyze the plant seismic response models to determine seismic initiated accident sequences and their frequency. This step involves the assembly and quantification of the plant logic models as well as the reporting and analyzing of the results.
: 10. Identify plant seismic vulnerabilities. This step defines any site specific vulnerabilities which are discovered as a result of the performance of the study. While the systematic process described above was used to identify and quantify seismic risk, SPRAs include major sources of uncertainty, as described in Aggregation of Quantitative Risk Assessment Results (EPRI 2005). The areas of uncertainty were summarized in that document
 
as follows:
* Hazard Curve: The seismic hazard curve is developed using a combination of actual data and expert judgment. The actual data used to develop the seismic hazard curve is generally very sparse. The expert judgment is generated using expert elicitation process and includes technical experts in their subject matter fields. However, technical experts tend to be conservatively biased as a result of a desire to be conservative knowing the implications of
 
the development of the seismic hazard curves is the design specifications for important safety systems. This conservatism is evidenced in the development of the distribution assigned to the hazard curve. With a larger distribution, the mean values of the frequency of occurrence increase.
* Fragility Curves: Fragility analysis performed in a typical seismic PRA is based on the "weak link" method. In this method, a seismic capacity engineer determines the weak link associated with a system or a particular function of a system, structure or component and develops a fragility of the component based on seismic acceleration. Similar to the Environmental Report Appendix E  SAMA ANALYSIS Page E-66 Three Mile Island Nuclear Station Unit 1 License Renewal Application development of a hazard curve, a combination of actual experience, testing, analysis, and expert judgment (to a lesser degree) is used to develop the fragility. The determination of the weak link is based on the subjective judgment of the seismic capacity engineer as is the final fragility albeit to a lesser degree.
* Correlation of seismic failures: Typical seismic PRA assume that systems, structures and components (SSC) that are similar are assigned a 100 percent failure correlation in the model. That is, one fragility applies to the failure of all similar components. For example, if a high pressure ECCS pump fails during a given seismic acceleration, then all similar ECCS pumps also fail. However, it is more likely that these components are not 100 percent correlated and that subtle, and sometimes not so subtle, differences between the components and their respective anchorages provide significant margins between the failure accelerations.
* Treatment of offsite power: In a typical seismic PRA a loss of offsite power is assumed for seismic events of any significant magnitude. The probability of a seismically induced loss of offsite power event can vary significantly and considerable judgment is usually used in the development of the fragility of the offsite power grid. In addition, the loss of offsite power is typically a significant contributor to the results of the seismic PRA.
* Treatment of balance of plant equipment: In a typical seismic PRA, the balance of plant equipment is omitted from the analysis as an analysis simplification. The reduction in the scope of the seismic PRA by the elimination of balance of plant equipment is performed to reduce the resources required to develop the seismic PRA. Generally, the balance of plant equipment is not seismically designed and details of the design and anchorage of the equipment is difficult to obtain, which further complicates the development of fragilities.
However, for some plant designs, specifically Boiling Water Reactors (BWRs), the balance of the plant systems provide significant mitigative potential. This is particularly true for the lower seismic accelerations where continued equipment operability is reasonably likely.
* Modeling simplifications: Other modeling simplifications are also employed to reduce the scope of the seismic analysis. These analysis simplifications are generally performed to reduce the scope of the fragility analysis which is resource intensive. These analysis simplifications include the treatment of human reliability analysis, support system operability/availability following a seismic event and others.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-67 License Renewal Application These characteristics of the SPRA limit the use of the absolute risk metrics that are a result of the analysis, but the relative rankings of the seismic contributors and the insights from the model are considered to be useful for identifying potential areas for plant improvements. E.5.1.6.2.2  Seismic Contributor Review For both the EPRI NP-6395-D and the NUREG-1488 seismic hazard curves, the largest CDF contributions came from the seismic events between 0.2g and 0.5g. The lower magnitude events (0.052g to 0.2g) had higher frequencies of occurrence, but the consequential damage to the plant systems was not severe and the conditional core damage probability was relatively low. The higher magnitude events (0.5g to 1.01g) caused heavy damage and resulted in high conditional core damage probabilities, but the frequencies of occurrence for seismic events of this magnitude were estimated to be low. The table below summarizes the Seismic CDF by initiating event category for both the EPRI and LLNL seismic hazard curves: TMI-1 Seismic Results Summary Based on EPRI NP-6395-D  Based on NUREG-1488 Initiating Event Earthquake Range  CDF Percent of Total CDF CDF  Percent of Total CDF SEIS1 0.052g to 0.2g 5.78E-06 18.0% 1.26E-05 14.9% SEIS2 0.2g to 0.3g 1.04E-05 32.4% 2.61E-05 31.0% SEIS3 0.3g to 0.5g 1.22E-05 38.0% 3.25E-05 38.6% SEIS4 0.5g to 1.01g 3.71E-06 11.6% 1.31E-05 15.5% As shown in the table above, the distribution of CDF among the initiating event categories remains consistent whether the EPRI or LLNL seismic hazard curves are used. The use of the LLNL seismic hazard curves amounts to a fairly linear increase in the CDF for each of the seismic initiating event categories without significantly changing the types of challenges that have the highest frequencies of occurrence. Because the absolute seismic CDF estimates are not directly used in the SAMA analysis, the choice of which seismic hazard curve is used to extract risk insights does not impact the SAMA analysis. Examination of the seismic component Fussell-Vesely values for the top contributors confirms this assertion:
Environmental Report Appendix E  SAMA ANALYSIS Page E-68 Three Mile Island Nuclear Station Unit 1 License Renewal Application Top Seismic Component Fussell-Vesely Contribution Summary Component ID Top Event Fussell-VeselyContribution (EPRI) Fussell-VeselyContribution (LLNL) Description FRAG15 GW 4.42E-01 4.00E-01 1P, 1S, 1R, 1T 480V Class 1E load centers seismic failure with offsite power available. FRAG15 GY 1.57E-01 1.50E-01 1P, 1S, 1R, 1T 480V Class 1E load centers seismic failure with offsite power failure. FRAG01 OX 1.21E-01 1.10E-01 Seismic offsite power insulator failure. FRAG20 CX 7.34E-02 7.00E-02 Seismic control room ceiling failure. FRAG09 GY 5.90E-02 6.00E-02 Seismic failure of EDG air start receivers. FRAG11 RX 2.05E-02 2.00E-02 Seismic failure of DHCCW heat exchangers. FRAG17 GY 1.07E-02 1.00E-02 Seismic failure of EDG ground resistors. A review of the LLNL based results shows that the largest Fussell-Vesely value for a non-seismic failure is Riskman top event "GA" (Class 1E AC power train "A") at 7.33 E-03, which implies that seismically induced failures are the main contributors to the seismic risk profile. As a result, the focus of the seismic review is based on the seismically induced failures rather than the independent failures. A review of each of the top seismic contributors is provided below.
FRAG15 This seismic component group includes 480V AC Class 1E load centers 1P, 1S, 1R, and 1T, which have been identified as components with low seismic ruggedness. As these load centers provide power to critical equipment and have HCLPF capacities slightly greater than the weaker off-site power related components, the availability of the load centers is important in both cases when off-site power is available and when it has failed. This is significant as seismic events that do not fail off-site power would not present a large threat to the site if the 480V AC load centers remained available. The low HCLPF values associated with these load centers demonstrate that the probability of a seismically induced failure of the equipment is not unlikely in earthquakes where off-site power remain s available, which is problematic. The IPEEE indicates that plant specific HCLPF values were estimated for these components types and were determined to be 0.12g. This implies that damage to these load centers may occur for even the SEIS1 initiating event group. One of the recommendations resulting from the IPEEE analysis was to reinforce the load center framework to prevent failure in seismic events; however, no work was done to strengthen the load center supports because the reduction in risk Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-69 License Renewal Application was considered to be low compared with the total CDF. Because these load centers are large contributors to the seismic risk profile and have not been strengthened since the IPEEE, reinforcing these 480V load centers has been added to the TMI-1 SAMA list (SAMA 27).
FRAG01 The ceramic insulators on the off-site power lines outside of the site and coming into the TMI-1 switchyard are susceptible to relatively low seismic shocks (HCLPF = 0.09). Other components, such as the auxiliary transformers and the 6.9kV AC distribution buses were also assessed to have the same low HCLPF capacity as the ceramic insulators. As a result, off-site power may be failed in many of the higher frequency, low magnitude earthquakes. The seismically induced LOOP requires the availability of the on-site AC systems to prevent core damage in the long term as recovery of off-site power is not credited in seismic events where widespread damage to the off-site AC distribution system could exist. Improving the off-site AC distribution system is not considered to be feasible and it is not suggested as a SAMA. Even if a seismically rugged, dedicated line to another generating station could be established, no information is available related to how other generating stations would respond to seismic challenges and their availabilities can not be assured. A more cost effective means of addressing the loss of off-site power cases would be to improve on-site AC power reliability. The issues related to improving the seismic capacities of plant components related to on-site AC power generation are discussed for FRAG20, FRAG09, and FRAG17 below. Another potential means of addressing off-site AC power failures is to implement changes that would allow the plant to operate without 4.16kV AC power for extended periods of time. As
 
described in Section E.5.1.6.1 , installation of the Westingho u se type high temperature, damage resistant seals would maintain primary side integrity while providing power to a 125V DC battery charger would allow for long term operation of the TD EFW pump for secondary side heat removal (SAMA 2). Even though control of the TD EFW pump is possible, powering the batter chargers is considered to be required because the 125V DC system is supports the vital 120V AC power supply for the OTSG level indicators. No credit is taken for operation of the TD EFW
 
pump without level indication.
Environmental Report Appendix E  SAMA ANALYSIS Page E-70 Three Mile Island Nuclear Station Unit 1 License Renewal Application FRAG20 Failure of the main control room ceiling is assumed to result in the loss of the "B" division of Class 1E AC power in the IPEEE. The consequences of the failure of the main control room ceiling are not highly predictable and may result in damage to other equipment, cause fires, injure plant operators, or on the other extreme, cause no damage at all. However, because these types of failures have the potential to impact important plant functions, supports were added to the main control room ceiling to reduce the likelihood of failure, as suggested in the IPEEE. The changes were accepted as adequate to address the identified issue and no additional changes are considered to be required to address control room ceiling failure.
FRAG09 The IPEEE identified a potential plant enhancement related to securing the EDG air start receivers to reduce the probability that they will fail after a seismic event. The changes were accepted as adequate to address the identified issue and no additional changes are considered to be required to address the EDG air start receiver anchorages.
FRAG11 The IPEEE identified a potential plant enhancement related to strengthening the anchorage used to secure the decay heat closed cooling water heat exchangers (DC-C-2A(B)) to reduce the probability that they will fail during a seismic event. This suggested change was reviewed, but not implemented as it was not considered to be a cost beneficial change. Failure of the DHCCW heat exchangers results in the loss of the ability to remove decay heat from the RPV and would lead to core damage if the secondary side heat removal function were also disabled. Given the low HCLPF capacity estimated for these components (0.09g) and the high importance of the DHCCW system, the anchorage enhancements suggested in the IPEEE have been included on the TMI-1 SAMA list for evaluation (SAMA 28).
FRAG17 Failure of the EDG ground resistors results in failure of the EDGs, which will lead to core damage in the event that off-site power is not available. Given that the HCLPF capacity for these components was estimated at 0.25g compared with 0.09g capacities of off-site power components (such as the 1/A and 1/B distribution buses or the aux transformers), it is likely that Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-71 License Renewal Application core damage will ensue due to long term loss of power if the EDG ground resistors fail from seismic shock. A potential means of addressing this issue would be to replace these components with a more seismically durable design (SAMA 29).
Diesel Fire Pump The IPEEE includes a potential plant improvement that suggests enhancing the supports for the diesel driven fire pump fuel oil tanks and batteries. This insight was based on a walkdown of the fire suppression system that was performed as part of the seismic/fire interaction assessment (not for the SPRA model). No quantitative estimates of seismically induced fire risk were presented in the IPEEE, but the conclusion based on the plant review was that the risk was low. However, this modification was included in the IPEEE as a potential plant
 
improvement given that the fuel tanks and battery racks appeared to have low seismic capacities and that the fire protection function could be degraded in a seismic event due to the weakness of the identified support structures. The supports for fuel oil tanks and batteries coul d be improved, but the impact of implementing these changes would be difficult to determine given that the SPRA assumed that the fire protection system was failed. The available results do not provide any insights on how improving the fire protection system's availability could impact risk. Based on the information in the current PRA model documentation, it is known that the fire protection system supports operation of the following equipment:
* SBO EDG engine cooling
* Backup cooling for the DHCCW heat exchangers (not credited)
* Backup Instrument Air 1A and 1B compressor cooling The SBO EDGs depend on the fire protection system as the primary engine cooling source. If the proposed fire protection system modifications were implemented, the fire protection system could be used to cool the SBO EDG in a seismi c event. However, because of the similarity between the "1E" EDGs and the SBO EDG, the seismic model would assume SBO EDG failure in the same scenarios where the 1E EDGs fail. The only likely benefit would come from cases Environmental Report Appendix E  SAMA ANALYSIS Page E-72 Three Mile Island Nuclear Station Unit 1 License Renewal Application where random failures disable the other two EDGs, which are much smaller contributors than other, seismic based equipment failures. The ability to provide backup cooling to the DHCCW system is of limited importance as the DHCCW heat exchangers, even with the improved anchorages, are the likely failure points of the system. In addition, the DHCCW system and the DHR system it supports depends on the availability of the AC distribution system, which may not be available. The 1A and 1B Instrument Air (IA) compressors are normally cooled by SSCCW, but fire protection is available as an alternate means of cooling in the event that SSCCW is unavailable. As the SSCCW heat exchangers are identified as low capacity components, the SSCCW system would likely be unavailable in even the 0.052g to 0.2g initiating event category causing failure of the IA system. Some improvement in the Instrument Air system availability could be gained through improving the fire protection system's seismic durability. While the Fussell-Vesely importance value for the IA system is low (0.01), improving the supports for the diesel fire pump fuel oil tank and the battery racks has been added to the
 
SAMA list (SAMA 30) to address p otential IA improv e m en t s, as iden t ified in Secti o n E.5.1.5. E.5.1.6.2.3 Containment Performance Analysis The effect of seismic events on the containment building performance was evaluated from two perspectives:
* Containment structure seismic capacity,
* Fragility of containment isolation valves and signals. The containment structure analysis concluded that the lowest median acceleration capacity for the containment building was 11.0g and that the HCLPF was 3.5g. Based on the high seismic capacity of the containment structure, seismic failure was not considered to be a credible event and no further evaluation was performed. As a result, no SAMAs are considered to be required to address containment building failures. The IPEEE analysis of the containment isolation function showed that most containment isolation valves would fail closed on loss of Instrument Air, which is a non-seismic designed system. As the lowest fragility of the containment isolation system was determined to be the Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-73 License Renewal Application ESAS relays at 0.89g, Instrument Air is not expected to be available after seismic events that challenge the containment isolation system components. One issue was identified related to the potential seismic interaction between containment purge line isolation valve AH-V-1D and
 
air supply tank PP-T-1A, which has a low seismic capacity. As a result of the IPEEE, the restraints for PP-T-1A were improved and failure of valve AH-V-1D due to contact with the tank was no longer considered to be an issue. No additional changes are suggested to address this issue. The only other containment isolation issue of concern was for motor operated valves that would fail "as-is" on loss of the corresponding power supply. The IPEEE concluded that the recovery times for containment isolation failure allowed sufficient time for manual or automatic closure of the valves after the seismic event, prior to core damage. No changes were considered to be required to address any of these types of release pathways in the IPEEE. While manual isolation is a proceduralized action at TMI-1 and is considered to be a credible recovery path for seismic scenarios, the containment penetration isolation valves were reviewed again for the SAMA analysis. In all cases where MOVs are used in containment isolation paths, it was determined that they are either on closed cooling system lines that would not provide a release path without additional failures, are on lines with diameters of one inch or less (not significant release paths), or are in series with AOVs and SOVs that would fail closed on loss of
 
air/power. The small containment penetrations (1 inch in diameter or less) do not provide a significant release pathway even if they are not isolated. These penetrations are screened from further review based on the small potential for release in conjunction with the ability to manually isolate
 
the valves, if required. The pathways that include MOVs in series with AOVs or SOVs that "fail closed" on loss of power/air are screened from consideration as the pathway would be isolated in loss of power cases that fail the MOVs. Manual action is also available to isolate the penetration in the event
 
that the "in series" AOV or SOV fails to close. Closed loop cooling systems could provide a release path through a "failed open" motor operated isolation valve; however, multiple boundary failures would be required in conjunction with core damage. For TMI-1, two closed loop cooling water system penetrations have been identified that include only MOVs as isolation valves:
Environmental Report Appendix E  SAMA ANALYSIS Page E-74 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Nuclear Services Closed Cooling Water: Three MOVs, NS-V-4, NS-V-15, and NS-V-35, are used as isolation valves on an 8 inch line which penetrates the reactor building.
* Reactor Building Normal Cooling Water: Two MOVs, RB-V-2A and RB-V-7, are used as isolation valves on 8 inch cooling lines which carry water to and from the reactor building cooling units. None of these penetrations are connected to the RCS and a release through either of these paths would require a pressurized containment atmosphere, a break in the reactor building side of the closed cooling water system boundary, and a break in the ex-reactor building side of the closed cooling water system boundary. These may be unlikely events, but no assessment of the probability of seismically induced failure of these pathways is available. The identified pathways could be isolated without operator actions if the valves were modified so that they fail in the "closed" position (SAMA 31). This SAMA is included the SAMA list, but it should be noted that changing the valves to "fail closed" introduces a failure mode for the valves that did not previously exist and may be detrimental in other accident scenarios. E.5.1.6.2.4 Seismic SAMA Identification Summary Based on the review of the TMI-1 seismic analysis, five Seismic related SAMAs have been identified:
* Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for Extended EFW Operation (SAMA 2),
* Improve the 480V AC load center welds (SAMA 27),
* Improve the DHCCW Heat Exchanger (DC-C-2A(B)) Anchorages (SAMA 28),
* Replace EDG Ground Resistors (SAMA 29).
* Improve Diesel Fire Pump Fuel Oil Tank and Battery Rack Supports (SAMA 30)
* Modify Specific Containment Penetration MOVs to "Fail Closed" (SAMA 31) E.5.1.6.3 High Wind Events The strategy taken to examine high wind risk in the TMI-1 IPEEE was to quantify the CDF due to high wind events and show that was below the screening frequency of 1.0E-06/yr. For the Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-75 License Renewal Application IPEEE, initiating events with a CDF below the screening frequency were precluded from further analysis and no detailed review of the plant response for these types of events was required. For TMI-1, the high wind based CDF (sum of high wind damage and missile strikes) was estimated to be 7.77E-07/yr based on some simplifying assumptions, including:
* The exceedance frequency used for 400 mph winds was taken to be the exceedance frequency for the 318-380 windspeed range (5.0E-04). The 400 mph wind speed was used to determine the tornado strike frequency because it was assumed to be the wind speed at which damage to category 1 structures could occur. This was based on the design limit of 360 mph and consideration of material stress safety factors employed in the design process. As  a result, the initiating event frequency may be overestimated,
* Any site tornado strike with wind speeds  400 mph is assumed to fail the BWST and CST and lead to core damage,
* Any tornado missile strike to outside equipment is assumed to fail the equipment. No potential plant improvements related to high wind risk were identified in the IPEEE as the events were screened from detailed analysis based on low frequency of occurrence. For the purposes of the SAMA analysis, an estimate of the cost-risk corresponding to high winds can be used to determine if any cost beneficial changes could be identified for the site. The cost-risk corresponding to high wind events is determined using the following assumptions:
* Internal and external events risk are approximately equal (excluding external flooding),
* The external events CDFs are directly proporti onal to the cost-risk associated with a given external event. For TMI-1, the internal events maximum averted cost-risk is $3,271,711, which implies that the non-external flood based external events contribution is also $3,271,711. For any given external event type, the corresponding cost-risk can then calculated by multiplying the total external event cost-risk by the ratio of the specific external event CDF to the total external events CDF (excluding external flooding). For example, for seismic events: seismic cost-risk = total external events cost-risk * (seismic CDF / total external events CDF) seismic cost-risk = $3,271,711 * (8.43E-05 / 1.07E-04) = $2,577,454 Environmental Report Appendix E  SAMA ANALYSIS Page E-76 Three Mile Island Nuclear Station Unit 1 License Renewal Application The following table summarizes the results for the non-flooding external events: External Events Cost-Risk Summary External Event CDF Ratio of CDF to TotalExternal Event CDF CorrespondingCost-Risk 3 Seismic 1 8.43E-05 7.88E-01 $2,577,454 Fire 2.16E-05 2.02E-01 $660,886 High Winds 7.77E-07 7.26E-03 $23,753 Aircraft Impact 2 3.95E-07 3.69E-03 $12,073 Hazardous Chemicals 1.60E-07 1.50E-03 $4,908 1 Based on the NUREG-1488 seismic hazard curves.
2 Intentional aircraft impact is treated outside of SAMA and is not accounted for here due to the specific nature of the threat. The CDF quantified in the IPEEE is used to address the potential for accidental impact. 3 T hese cost-risks are calculated by m u ltipl y ing t he e x ternal events ba s ed c o s t-risk (see section E.4.6) by the percent contribution of the external event type. The cost-risk associated with high winds is only $23,753, which is less than the minimum expected cost of implementation of
$50,000 (see section E.5
.1). As a result, it is unlikely that any cost-beneficial SAMAs could be found to reduce the risk of high wind events and no further review is considered to be required for the SAMA analysis. E.5.1.6.4 External Flooding As part of the TMI-1 IPEEE, the site was reviewed to identify the largest flooding risks. This included high river flows from dam breaks, hurricane effects, snow melt, and other non-hurricane events. The bounding risk was determined to be a flood of the Susquehanna River
 
most likely caused by a hurricane event. The external flooding analysis performed in the TMI-1 IPEEE divided flood risk into three categories:
* Floods with elevations greater than 310 feet mean sea level (msl)
* Floods with elevations between 305 and 310 feet msl,
* Floods with elevations less than 305 feet msl. The main contributors to core damage for each of these flood elevation ranges are different and are examined separately for the SAMA analysis.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-77 License Renewal Application E.5.1.6.4.1 Floods Great er than 310 Feet msl Given the configuration of the plant at the time of the IPEEE, floods with elevations over 310 feet msl were assumed to result in the loss of all electrical equipment due to flooding of site buildings. As the existing flood gates would not prevent flooding of these buildings for these scenarios, successful installation of the flood gates would increase the length of time available before building flooding, but not prevent core damage. Based on insights from the IPEEE and previous TMI-1 external flooding analyses, a strategy was implemented at the site to use a temporary power source and submersible pumps to maintain the reactor in a safe state during these extreme flood conditions. The CDF of 8.10E-05/yr that was reported in the IPEEE credited the use of this strategy. As is the case with the other external events contributors, the level of development and uncertainty of the external flooding results is not comparable to the current internal events PRA. Assuming that external flooding risk dominates the risk profile for TMI-1 because the CDF is about two times greater than the internal events CDF is not necessarily correct. However, because there are no reliable means of demonstrating that floods exceeding 310 feet msl are low risk events and because the consequences of the events are severe, TMI-1 should have a reliable method in place to address these scenarios. Tangible work has already been completed at TMI-1 to satisfy this need, but the flood scenarios must be considered in the context of the SAMA analysis to determine if additional changes could be cost-beneficial. Based on the evaluation presented in the TMI-1 IPEEE, the major contributors to the CDF for flood events over 310 feet msl include:
* Failure of secondary side cooling (7.03E-02) (represented only by operator error),
* Failure of primary side makeup and seal injection (4.22E-02) (represented only by operator error).
* Failure of the portable EDG in the 48 hour mission time (1.43E-01) These contributors can be evaluated to identify areas of weakness and potential means of improving the associated reliabilities. The guidance that was developed to mitigate floods greater than 310 feet msl as a result of the IPEEE is considered to provide an appropriate level of detail for the actions required in the Environmental Report Appendix E  SAMA ANALYSIS Page E-78 Three Mile Island Nuclear Station Unit 1 License Renewal Application relevant scenario. For the current configuration, no additional risk reduction is considered to be possible through procedural changes alone. Flood risk could be reduced by improving the state of readiness of the corresponding equipment (prestaging, SAMA 32). Examples of the things that should be considered include:
* Permanently mount the power cables between the generator and pump staging areas,
* Permanently mount injection lines required for primary and secondary side makeup (may not be practical for the secondary side pump that takes suction from flood water in the turbine building),
* Consider an alternate secondary side suction source given that flood waters may recede well before an alternate secondary side makeup source will become available when AC power is re-established to the site,
* Ensure the power cables have all required connectors attached or stored in the staging areas,
* Pre-manufacture any required air supply or fuel oil connectors and store them in the staging areas,
* Stage the portable generator on the turbine deck or provide a means of hoisting the generator and fuel oil to the turbine deck when offsite power is not available, Another area of interest is the reliability of the portable EDG. The operation of the portable diesel generator for the 48 hour mission time is a large contributor to failure that is based on data similar to what is used in the PRA. While it may be true that the failure rate for the portable generator is much less than a standard EDG, a lower failure rate cannot be justified without a verifiable data source. A potential means of improv ing the reliability of the temporary AC supply would be to procure a spare 480V AC generator.
An alternative to the pre-staging option would be to increase the flood height for which the unit is protected (SAMA 33). The current configuration protects to the design basis limit of 310 feet msl and levels any higher result in topping of the existing flood doors and flooding of sensitive areas. In order to decrease the flood CDF to about 1E-5/yr, the flood protection height would Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-79 License Renewal Application have to be increased to 324.5 feet msl on the following gates/structures (completely sealing doors is suggested, where possible):
EDG Building
* Gate D-1
* Gate D-3
* Gate D-4
* Air Vent Valves for the fuel oil day tanks
* Seal underground cable vaults to prevent short circuits due to water incursion Air Intake Structure
* Access Door
* Air Intake Vents Intermediate Building
* Gate C-1 Control Building
* Gate B-1
* Gate B-2 Intake Screen Pumphouse
* Gate E-1
* Gates E-2
* Gate E-3
* Gate E-4 Environmental Report Appendix E  SAMA ANALYSIS Page E-80 Three Mile Island Nuclear Station Unit 1 License Renewal Application By preventing the incursion of water, the existing safety equipment should be capable of maintaining safe shutdown conditions as long as fuel oil is available to the EDGs. E.5.1.6.4.2 Floods with Elevations Between 305 and 310 Feet msl Flood events with elevations between 305 and 310 feet msl were evaluated with an event tree in order to describe and quantify the various core damage scenarios initiated by such floods. Of the six core damage scenarios evaluated in the event tree, three scenarios contributed 94 percent of the risk:
* Sequence CD-A (36.8%): Flood frequency (305 to 310' msl)
* probability off-site power is available
* probability of failing to install flood gates (cold shutdown achieved)
* probability of failing to implement severe flood cooling,
* Sequence CD-D (35.7%): Flood frequency (305 to 310' msl)
* probability off-site power is unavailable
* probability that cold shutdown is not achieved prior to flood (off-site power not available)
* probability of failing to install flood gates (cold shutdown not achieved)
* probability of failing to implement severe flood cooling,
* Sequence CD-E (21.4%): Flood frequency (305 to 310' msl)
* probability off-site power is unavailable
* probability of on-site power failure
* probability of failing to implement severe flood cooling. A common failure of sequences CD-A and CD-D is the inability to implement the severe flooding cooling strategy that was designed for floods over 310 feet msl. While this cooling strategy was intended to mitigate only the most severe floods, it can be used for any flooding events where Turbine Building flooding occurs given that t he secondary side submersible pump uses the flood water in the Turbine Building as a suction source (the primary cooling pump suction source is the SFP and it is potentially available for any condition). For floods with elevations between 305 and 310 feet msl, damage to plant safety equipment requires failure of the flood gates. This requirement implies that the time available to implement the severe flooding cooling strategy is less than for the scenarios where flood elevations must rise to greater than 310 feet msl. As a result, the human error probability associated with this action is larger than for the scenarios with flood elevations over 310 feet msl. The IPEEE assumed an HEP of 5.0E-01 for implementing the severe flooding cooling alignment for the 305 to 310 foot msl floods. The improvements to the severe flooding mitigation strategy suggested for floods greater than 310 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-81 License Renewal Application feet msl (SAMA 32) would also reduce the human error probability for sequences CD-A and CD-D and is considered to be an effective SAMA for these flood events. Another common failure between sequences CD-A and CD-D is related to flood gate installation. The IPEEE assessment concluded that human error was the dominant factor related to flood gate installation failure and only human error was included in the failure probability. The HEPs for flood gate installation failure used in the IPEEE ranged from 5.6E-02 to 6.3E-02 based on the contemporary flood gate design. Since that evaluation, TMI-1 replaced the seal system on the Unit 1 class 1 buildings. The changes were considered to have improved seal reliability, made the seals easier to maintain, and made the seals/gates more convenient to use. These changes may have improved the reliability of flood gate installation in some way; however, the HEPs were not re-quantified to reflect the gate enhancements. Further changes to the gates could be made to improve their ease of use, such as replacing all gates with permanently installed swinging gates that could be secured with a handwheel. While such a change may make the flood gates easier to use, it would be difficult to justify a large difference in the HEP associated with the improved gate system and the current design given the long period of time that is available to properly install the gates. Based on an onsite review of the gates and discussions with the flooding engineer, no changes to the gates are suggested to improve their installation mechanisms. Sequence CD-E appears to be a simplified evaluation of the cases in which both on-site and off-site AC power fail. From the information in the IPEEE submittal, the flooding event tree shows that core damage occurs for floods between 305 and 310 feet msl elevation if off-site power fails in conjunction with on-site AC power. However, the total CDF from the event tree is multiplied by the 0.5 failure probability for severe flooding cooling alignment to obtain the final CDF for floods with elevations between 305 feet msl and 310 feet msl. This implies that core damage does not occur before flood waters reach a level where the submersible pump could be used for secondary side cooling. No discussion is provided to describe the timing of off-site power loss relative to the flood height. This is important because the accident would evolve differently depending on whether it is caused by a hurricane or by flooding of the site transformers. For the case of a hurricane induced flood, offsite power could be lost early and on-site power would therefore be challenged at that time. An early failure in on-site power would result in core damage before flood water reaches the turbine building and no credit should be taken for the existing severe flooding cooling alignment. Quantitative resolution of this issue would require a more detailed analysis than what was performed for the IPEEE, but this uncertainty could be Environmental Report Appendix E  SAMA ANALYSIS Page E-82 Three Mile Island Nuclear Station Unit 1 License Renewal Application addressed through implementation the alternate se condary side suction source that is part of the SAMA 11 design. This provides a means of initiating both primary and secondary side makeup at any time during an accident. A lower frequency sequence, CD-F, is based on the failure to provide an early flood warning to the plant. The early flood warning was assumed to be the cue instigating the initiation of the required flood procedures. While no credit was taken for installing the flood gates in time to prevent flooding of site buildings, the IPEEE credited implementation of the severe flooding cooling alignment. No discussion was identified that described why credit was taken for the severe flooding cooling alignment under this circumstance, but sequence CD-F would only comprise about 1 percent of the external flooding CDF if credit for the alignment were disallowed. Due to the low contribution of sequence CD-F relative to the other sequences, no SAMAs are considered to be required. E.5.1.6.4.3 Floods with Elevat ions Below 305 Feet msl In order for floods in this category to impact the site, the dike on the northern tip of the island is required to fail in conjunction with the flood event. The frequency of these events, which are required to cause site flooding, were determined to be less than 3 percent of the total flooding frequency alone. The conditional CDF given this type of flood event was estimated to be less than 0.1, which would correspond to a contribution of less than 0.3 percent of the total external flooding CDF. No detailed CDF sequences were developed for these floods in the IPEEE and no specific failure contributions were identified other than dike failure. Improvements could be made to the dike, but even with dike failure, the buildings housing safety equipment would not flood. The only potential issue identified is the flooding of the EDG building cable vaults, which is addressed by SAMA 33. No other SAMAs are suggested to address this flood category. E.5.1.6.4.4 External Flooding SAMA Identification Summary Based on the review of the TMI-1 external flooding analysis, two external flooding related SAMAs have been identified:
* Prestage Severe Flooding Equipment (SAMA 32),
* Increase the Flood Protection Height (SAMA 33)
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-83 License Renewal Application E.5.1.6.5 Transportation and Nearby Facility Accidents Transportation and nearby facility accidents were included in the TMI-1 IPEEE to account for human errors or equipment failures that may occur in events not directly related to the power generation process at the plant. The types of hazards explicitly evaluated for the site include:
* Aircraft Impact
* Hazardous Chemical Release E.5.1.6.5.1 Accidental Aircraft Impact At the time the IPEEE was performed, available information related to military, commercial, and general aviation traffic was used to estimate the frequency of a release of radionuclides caused by aircraft impact. Given the information and conditions present at the time of the analysis, the frequency was determined to be 3.95E-07 per year and further analysis was not considered
 
warranted. It is recognized that the types of credible threats to nuclear facilities by aircraft have changed since the time the IPEEE was published. While this is true, efforts are underway within the industry to address this issue in conjunction with other forms of sabotage. Based on the fact that this topic is currently being analyzed in another forum and due to the complexity of the issue, intentional aircraft impact events are considered to be out of the scope of the SAMA analysis. The analysis performed in the IPEEE is used to provide insights related to accidental aircraft impact. No potential plant improvements related to the risk of accidental aircraft impacts were identified in the IPEEE as the events were screened from detailed analysis based on low frequency of occurrence. For the purposes of the SAMA analysis, an estimate of the cost-risk corresponding to accidental aircraft impact can be used to determine if any cost beneficial changes could be identified for the site. The cost-risk corresponding to accidental aircraft impacts is determined using the following assumptions:
* Internal and external events risk are approximately equal (excluding external flooding),
* The external events CDFs are directly proporti onal to the cost-risk associated with a given external event.
Environmental Report Appendix E  SAMA ANALYSIS Page E-84 Three Mile Island Nuclear Station Unit 1 License Renewal Application For TMI-1, the internal events maximum averted cost-risk is $3,271,711, which implies that the non-external flood based external events contribution is also $3,271,711. For any given external event type, the corresponding cost-risk can then calculated by multiplying the total external event cost-risk by the ratio of the specific external event CDF to the total external events CDF (excluding external flooding). For example, for seismic events: seismic cost-risk = total external events cost-risk * (seismic CDF / total external events CDF) seismic cost-risk = $3,271,711 * (8.43E-05 / 1.07E-04) = $2,577,454 The following table summarizes the results for the non-flooding external events: External Events Cost-Risk Summary External Event CDF Ratio of CDF to TotalExternal Event CDF CorrespondingCost-Risk 3 Seismic 1 8.43E-05 7.88E-01 $2,577,454 Fire 2.16E-05 2.02E-01 $660,886 High Winds 7.77E-07 7.26E-03 $23,753 Aircraft Impact 2 3.95E-07 3.69E-03 $12,073 Hazardous Chemicals 1.60E-07 1.50E-03 $4,908 1 Based on the NUREG-1488 seismic hazard curves.
2 Intentional aircraft impact is treated outside of SAMA and is not accounted for here due to the specific nature of the threat. The CDF quantified in the IPEEE is used to address the potential for accidental impact. 3 T hese cost-risks are calculated by m u ltipl y ing the e x ternal events ba s ed c o st-r i sk (see section E.4.6) by the percent contribution of the external event type. The cost-risk associated with aircraft impact is only $12,073, which is less than the minimum
 
expected cost of implementation of
$50,000 (see section E.5
.1). As a result, it is unlikely that any cost-beneficial SAMAs could be found to reduce the risk of accidental aircraft impact
 
events. It should be noted that the accidental aircraft impact assessment from the IPEEE was based on air traffic assumptions relevant to the initial license period. That assessment assumed a continuous "aircraft movement" growth for Harrisburg International Airport that was two to four
 
times larger than the national average growth observed for the years 1979 to 1988. This resulted in an estimate of 177,000 aircraft movements per year for the midpoint of the original license period (1994). In order for the minimum cost SAMA (a procedure change of $50,000) to be potentially cost effective, the aircraft movement frequency would have to increase by a factor of 4.5. Even an order of magnitude increase in the aircraft movement would only yield a Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-85 License Renewal Application potential averted cost-risk of $120,730 for a completely effective SAMA. Based on the small impact of large changes in aircraft activity, any changes to aircraft movement frequency that may occur over the license renewal period are not expected to increase accidental aircraft impact risk to the point where potential SAMAs would become cost-effective. No SAMAs are suggested to address accidental aircraft impact for TMI-1.
E.5.1.6.5.2 Accidental Hazardous Chemical Release Similar to the aircraft impact assessment performed for the IPEEE, the hazardous chemical release assessment was based on non-intentional events. Threats related to intentional chemical releases are credible; however, the specialized nature of security threats requires that they are treated in a separate forum and they are not addressed as part of the SAMA analysis. For accidental releases, the IPEEE considered stationary and transient hazardous chemical sources that could pose a threat to TMI-1 if a release were to occur. As shown in the accidental aircraft impact discussion above, the cost-risk associated with hazardous chemical releases is only $4,908 assuming that the conditions present at the time of the IPEEE are applicable.
Some variation may occur in the characteristics of the chemical loads near the site or transported on the rail lines close to the site over the course of the license renewal period. While it is not possible to accurately predict what these changes could be, an order of magnitude increase in the risk that was estimated in the IPEEE would only increase the associated cost-risk to $49,080. Given that an order of magnitude increase in the hazardous chemical release risk would still not be likely to yield any cost beneficial plant changes, no SAMAs are suggested to address these types of threats. E.5.2 PHASE I SCREENING The initial list of SAMA c andidates is presented in Table E.5-3. The process used to develop the initial list is desc r ib e d in Section E.5.1. 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:
* Applicability to the Plant:  If a proposed SAMA does not apply to the TMI-1 design, it is not retained.
Environmental Report Appendix E  SAMA ANALYSIS Page E-86 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Implementation Cost Greater than Screening 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.
Table E.5-3 provides a description of how each SAMA was disposition e d in the Phase I analysis. All SAMAs that were found to be applicable to the TMI-1 design and to have a cost of implementation less than the MACR were passed to the Phase II analysis for a more detailed evaluation (Section E.6
).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-87 License Renewal Application E.6 PHASE II SAMA ANALYSIS Not all of the Phase II SAMA candidates require detailed analysis. The Phase II process allows for the screening of SAMAs known to be related to non-risk significant systems or to components/functions with low importance rankings. Due to the nature of the PRA based process used to develop the TMI-1 SAMA list, there are limited avenues for SAMAs of this type to be included in the list. However, potential pathways do exist:
* Inclusion of unresolved proposed plant changes from previous TMI-1 risk analyses,
* Inclusion of SAMAs based on the results of conservative modeling methods. While no calculations are required for eliminating a SAMA that is linked to a non-risk significant system or components, some quantitative efforts are usually required to screen SAMAs that were developed to address risk contributors based on conservative modeling techniques.
These cas e s are identi f ied in Table E.5-4 and discussed in d etail in the S A MA specific subsections of E.6. For the SAMAs requiring detailed analysis, a more detailed conceptual design was prepared to allow the p r oposed SAMA to be modeled in t h e PRA. The results of t h e model changes were used in conjunction with the estimated implementation costs to evaluate whether or not the SAMA is cost beneficial. The final cost based screening method is defined by the following equation: Net Value = Averted cost-risk - cost of implementation Where: Averted cost-risk = (baseline maximum averted cost-risk - maximum averted cost-risk with SAMA implemented) If the net value of the SAMA is negative, the cost of implementation is larger than the benefit associated with the SAMA and the SAMA is not considered beneficial. The baseline MACR was
 
derived using the methodology presented in Section E.4. T h e MACR wi t h the SAMA implemented is determined in the same manner with the exception that the PRA results used as input reflect implementation of the SAMA.
Environmental Report Appendix E  SAMA ANALYSIS Page E-88 Three Mile Island Nuclear Station Unit 1 License Renewal Application The calculation of the averted cost-risk for a SAMA must account for external events contributions. In some cases, representing t he impact of a SAMA's impact on external events is complex. The method adopted in the SAMA analysis to address this issue is dependent on the
 
type of SAMA to be quantified:
* For SAMAs that were not specifically developed to address external events issues, the multiplier d e fined in Sec t ion E.4.6.3 is used on t h e internal events averted cost-risk to provide an estimate of the non-external flooding external events benefit. This serves only as a gross approximation of the true benefit given that a SAMA may not impact both internal and external events risk in the same way. The external flooding model is quantified
 
separately.
* For SAMAs that were specifically developed to address external events, the external events models are used to extract quantitative insights that can be used to provide bounding estimates of potential averted cost-risks. In these cases, the specific external events benefit calculations generally supercede the multiplier and the multiplier is not used. The details of the quantification process vary for each SAMA and are described in the SAMA specific discussions of Section E.6. The implementation costs used in the Phase II analysis include both TMI-1 specific estimates developed by plant personnel and estimates taken from other SAMA submittals for those SAMAs that were determined to be highly similar. It should be noted that the TMI-1 specific implementation costs do not specifically include contingency costs for unforeseen difficulties nor
 
do they account for any replacement power cos t s that may b e incurred d ue to cons e quential shutdown time.
Sections E
.6.1 - E.6.33 describe t h e detailed cost-benefit a nalysis that was used f o r each of the remaining candidates. E.6.1 SAMA NUMBER 1:  ENHANCE THE SBO EDG WITH AUTO START AND LOAD CAPABILITY The availability of an auto start and load function for the SBO EDG will reduce the time required to restore power to the RCP seal cooling systems when the AC power has been lost and the "A" and "B" EDGs fail. Procedures should be reviewed to ensure that they will allow the operators to establish at least one form of seal cooling within 13 minutes of the initial loss of cooling. This is critical given that restoring RCP seal cooling after the 13 minute limit is considered to cause Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-89 License Renewal Application damage to the seals that will exacerbate seal leakage. The benefit of this SAMA would be enhanced if the auto start/load logic were capable of backing up either division of power for single EDG failures and selecting a single division to support in the event that both the "A" and
 
"B" EDGs fail.
The SBO EDG is described in the plant manuals as being capable of accepting a load within 10 minutes of an SBO, but no credit is taken in the PRA for preventing seal damage due to the uncertainty in this performance time and the time required to ensure seal cooling is established.
Some additional margin may be possible through procedure optimization, but the time window for action is so short that the most reliable way ensuring seal cooling is re-established before the 13 minute limit is reached is through automation of the start and load process for the SBO EDG. E.6.1.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, it was necessary to perform both basic event data changes and event tree/fault tree structure modifications given that the LOOP-SBO event tree is structured to force a seal LOCA when the "A" and "B" EDGs are unavailable. Specifically, the operator action to start the SBO EDG was reduced by a factor of 10 to represent automation of the start function. Further, it was necessary to adjust the joint human error probabilities (JHEPs) that included the action to start the SBO EDG given that the manual start action is essentially eliminated by the SAMA. In this case, it is appropriate to eliminate all JHEPs associated with the SBO EDG start action as the automated start removes the human action from the dependence chain. With respect to the impact on seal LOCAs, the LOOP-SBO tree logic was changed to allow the SBO EDG to prevent a seal LOCA. The following table summarizes the model changes that were made:
Environmental Report Appendix E  SAMA ANALYSIS Page E-90 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 1 - Model Changes Gate and / or Basic Event ID and Description Description of Change GSHEO1A----HDGOA: O PERATOR FAILS TO START SBO DG The basic event probability was changed from 2.66E-02 to 2.66E-03. JHHNSHOTHEOHEPOA: JHHNS10HOT1HEPOA AND GSHEO1A----HDGOA (dependence with tripping the RCPs) The basic event probability was changed from 3.60E-05 to 0.0. JHHNS10HEO1HEPOA: NRHNS10_HERHP1OA AND GSHEO1A----HDGOA (dependence with restarting NSRW pumps after a loop) The basic event probability was changed from 3.10E-04 to 0.0. LOOP-030 through LOOP-052 These gates were removed from the model as they are no longer required. The gates were previously used to model sequences in which a seal LOCA developed when only the SBO EDG was available. RCP-LOOP-100 This gate was removed from the model as it was previously used to delineate cases where only the SBO EDG was available. These cases would previously result in seal LOCAs, but the SAMA implementation eliminates this condition. It should be noted that the modeling strategy outlined above conservatively forces SAMA 1 to mitigate all seal LOCA cases with successful EFW operation when 4kV AC power has been lost to a single AC bus. There are scenarios in which core damage will occur for these conditions with SAMA 1 in place, but the impact is minor and would not change the conclusions for this SAMA. The results of the quantification are summarized below:  SAMA 1 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 1.88E-05 27.51 $98,718 Percent Change
-20.7% -15.6% -12.1% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-91 License Renewal Application SAMA 1 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-11 9.07E-11 1.90E-10 2.88E-10 3.90E-08 1.46E-08 8.54E-09 3.16E-07 7.39E-07 1.66E-07 2.20E-08Freq. (/yr)SAMA 4.56E-07 1.59E-06 1.81E-07 1.27E-08 4.21E-11 4.21E-11 1.90E-10 2.40E-10 1.14E-08 1.24E-08 1.03E-09 2.88E-07 5.55E-07 1.34E-07 1.68E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.03 0.04 0.00 0.84 3.41 0.82 0.05 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,677 $44,202 $3,367 $236 $2 $2 $7 $9 $102 $111 $9 $2,589 $11,211 $2,707 $159 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-07 2.75E-09 7.45E-07 2.89E-07 3.19E-06 1.32E-05 1.69E-08 2.36E-06 1.91E-08 2.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 7.00E-08 6.41E-09 1.68E-07 2.58E-09 6.48E-07 9.53E-08 2.21E-06 1.07E-05 1.66E-08 1.63E-06 1.59E-08 1.88E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.20 0.02 0.23 0.00 0.87 0.13 4.91 2.85 0.00 0.44 0.00 27.51 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $662 $61 $642 $10 $2,475 $364 $13,879 $2,798 $4 $427 $4 $98,718 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 1 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,790,086 $481,625 2.0 $963,250 E.6.1.2 EXTERNAL FLOODING EVALUATION This SAMA can have an impact on any scenario requiring the operation of the SBO EDG. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, the SBO EDG is flooded and this SAMA has no impact on the risk.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by the enhanced capabilities of the SBO EDG as core damage is caused by failure of the flood gates (the SBO EDG is flooded) or because a flood warning is not provided and no preparations are made for the flood (the SBO EDG is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of all AC power leads to core damage. Given that a loss of all power implies failure of the SBO EDG, SAMA 1 would provide no benefit to Sequence "E".
Environmental Report Appendix E  SAMA ANALYSIS Page E-92 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Floods below 305' msl:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. While the inclement weather conditions that would likely exist in flood scenarios would provide an indication that a LOOP may occur, the operators would not start and prepare the SBO EDG for loading before the onset of loss of AC conditions. As a result, these scenarios are assumed to be impacted in the same way as the internal events LOOP events are. In order to quantify the flooding benefits, it was necessary to characterize the impact of SAMA 1 on the internal events LOOP sequences. Once this is completed, the frequency of floods below 305' msl can be reduced by the same percentage. Because SAMA 1 predominantly impacts LOOP events, the absolute reduction in LOOP CDF can be calculated by subtracting the CDF for SAMA 1 from the base CDF: Absolute LOOP CDF Reduction = 2.37E 1.88E-05 = 4.90E-06 The total base LOOP CDF can be approximated by multiplying the Fussell-Vesely value of the LOOP initiating event (%AC) by the base CDF:
Base LOOP CDF = 3.26E-1
* 2.37E-05 = 7.72E-06 The percent reduction in the LOOP CDF can then easily be determined: Percent Reduction in LOOP CDF = (Absolute LOOP CDF Reduction / Base LOOP CDF)
* 100 Percent Reduction in LOOP CDF = (4.90E-06 / 7.72E-06)
* 100 = 63.5% Based on these results, the CDF for the floods below 305' msl was reduced by 63.5%. The following tables summarize the results of these changes: SAMA 1 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.09E-05 176.92 $541,385 Percent Change
-0.2% -0.1% -0.1% A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-93 License Renewal Application SAMA 1 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 9.13E-08 8.09E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.13 176.92 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $445 $541,385 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 1 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,520,578 $22,895 E.6.1.3 COST OF IMPLEMENTATION The cost of this SAMA was estimated to be $3,125,000 (Exelon 2007c). E.6.1.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 1 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $963,250 $22,895 $986,145 $3,125,000 -$2,138,855 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E  SAMA ANALYSIS Page E-94 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.2 SAMA NUMBER 2:  INST ALL DAMAGE RESISTANT HIGH TEMPERATURE RCP SEALS WITH A PORTABLE 480V AC GENERATOR FOR EXTENDED EFW OPERATION RCP seals have been developed that are capable of preventing seal LOCAs on loss of seal cooling events. The Flowserve N-9000 seals are reported to limit seal leakage to about 1 gpm per RCP seal even when cooling to the seals is completely lost, which is essentially considered to eliminate the seal LOCA evolution. In SBO cases, prevention of a seal LOCA will allow for extended operation if level instrumentation can be supplied using the vital 120V AC system.
Powering the station battery chargers with a portable 480V AC generator would provide this capability and allow control of the TD EFW system to be retained in the MCR. In order to maintain control of the TD EFW system from the MCR, power must be supplied for multiple loads, including:
* Level instrumentation,
* Control of EF-V-30 valves, and
* Instrument air for EF-V-30 valves. The 480V AC generator should be capable of providing these loads as long as the correct connections are made and the loads are managed properly. Cooling water is another concern for the instrument air compressors, but IA-P-1A and IA-P-1B can be cooled from the Altitude Tank. Plant documentation indicates that this connection is linked to the Fire Service system, so it is also assumed that Fire Service water could be used in an SBO based on the availability of the diesel driven fire pump. In the event that one of these support systems fails, it is also possible to operate the EF-V-30 valves locally, without any support other than power for SG level instrumentation. E.6.2.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-95 License Renewal Application averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate the installation of a new RCP seal package that prevents the onset of a RCP seal LOCA, a recovery event was appended to cutsets using QRECOVER32 that satisfied the gate logic for RCP-LOOP-100, "RCP SEAL FAILURE". This process captures all of the seal LOCAs contributors and multiplies them by the probability of the recovery event, which in this case was set to 1.0E-01. While the new seals may be capable of preventing a seal LOCA with a reliability greater than 90% when cooling is lost, the existing PRA model is not configured to analyze the probability of core damage after a seal LOCA is prevented. Ten percent of the original seal LOCA contribution is retained to represent:
* The CDF from cases where the new seals fail and a seal LOCA occurs,
* The CDF from cases where the new seals prevent a seal LOCA, but the core is damaged due to other failures. In order to account for the reduction in CDF due to the availability of a spare 480V AC diesel generator to supply backup 480V AC power, the HEP event EFHEF1_OPERH2HOA was reduced by a factor of 10. The CDF reduction is primarily due to the improved performance shaping factors related to the ability of the operator to use the MCR controls for EFW, but there may also be some improvement in the HEP related to the reduced manipulation time for the action. In the TMI-1 model, the relevant operator actions include the independent event discussed above as well as joint human error events. In this case, allowing for continued control of EFW in the MCR would not eliminate the dependence with other actions as the mechanism of dependence is primarily cognitive, but it could impact the JHEP probabilities.
Depending on the nature of the JHEP calculation, the actual impact on the JHEP probabilities could range from a percent or two all the way to a factor of 10. Rather than recalculate the JHEPs, they were conservatively eliminated for convenience. No HEP is included for failure to align the portable 480V AC generator. For this evaluation, it is assumed that the operators will always be able to align the charger before depletion of the batteries and that the generator will
 
always run. No model requantification was performed for this SAMA. All of these operations were performed on the existing base cutset files through basic event data changes and cutset Environmental Report Appendix E  SAMA ANALYSIS Page E-96 Three Mile Island Nuclear Station Unit 1 License Renewal Application recovery. The following table summarizes the changes that were made to the basic event data and a brief description of the recovery file used to modify the seal LOCA contributors: SAMA 2 - Model Changes Gate and / or Basic Event ID and Description Description of Change EFHEF1_OPERH2HOA: OPERATOR FAILS TO MANUALLY OPERATE EF-V-30 AFTER LOSS OF INSTRUMENT AIR The basic event probability was changed from 2.00E-03 to 2.00E-04. JHHEF1-HBW1HEPOA: EFHEF1_OPERH2HOA AND BWHBW1-----HP2OA (dependence between EF-V-30 operation and manual HPI initiation) The basic event probability was changed from 1.00E-04 to 0.0. JHHAM2-HEF1HEPOA: AMHAM2-----HC1OA AND EFHEF1_OPERH2HOA (dependence between EF-V-30 operation and manual start of air compressors after a LOOP) The basic event probability was changed from 4.61E-03 to 0.0. JHHAMHEFHBWHEPOA: JHHAM2-HEF1HEPOA AND BWHBW1-----HP2OA (dependence between EF-V-30 operation, manual start of air compressors after a LOOP, and manual initiation of HPI) The basic event probability was changed from 2.40E-04 to 0.0. JHHAM1-HEF1HEPOA: AMHAM1-----HC1OA AND EFHEF1_OPERH2HOA (dependence between EF-V-30 operation and failure to bypass IA dryer transfer valve) The basic event probability was changed from 1.81E-02 to 0.0. JHHAMHEFHB2HEPOA: JHHAM1-HEF1HEPOA AND BWHBW1-----HP2OA (dependence between EF-V-30 operation, failure to bypass IA dryer transfer valve, and manual initiation of HPI) The basic event probability was changed from 4.90E-05 to 0.0. RCP-SEAL-IMPROVE.CAF New recovery fault tree with top gate "Recoveries" used to apply a recovery event (RCP-SEAL-IMPROVE) to all cutsets including seal LOCAs.
The new gates include:
* RECOVERIES (Equivalence gate connected to new gate RCP-SEAL-
 
IMPROVE)
* RCP-SEAL-IMPROVE (Equivalence gate connected to existing gate RCP-LOOP-
 
100) Note that the action EFHEF2_OPERHFCOA and its JHEPs are not included in the model changes tabulated above as they have no measurable impact on the CDF. The results of the quantification are summarized below:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-97 License Renewal Application SAMA 2 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 1.13E-05 15.24 $56,521 Percent Change
-53.3% -53.3% -49.7% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 2 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.18E-07 9.21E-07 1.80E-07 9.32E-09 0.00E+000.00E+000.00E+000.00E+007.26E-091.25E-094.86E-11 2.76E-08 3.56E-074.98E-085.87E-09Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.39 5.27 0.91 0.05 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.08 2.19 0.31 0.02 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $11,620 $25,604 $3,348 $173 $0 $0 $0 $0 $65 $11 $0 $248 $7,191 $1,006 $55 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.43E-11 0.00E+00 6.57E-09 1.43E-09 6.77E-084.09E-107.20E-083.75E-086.73E-078.14E-065.45E-09 3.18E-07 8.19E-101.13E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.02 0.00 0.09 0.00 0.10 0.05 1.49 2.17 0.00 0.08 0.00 15.24 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $0 $0 $62 $14 $259 $2 $275 $143 $4,226 $2,133 $1 $83 $0 $56,521 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 2 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $1,595,737 $1,675,974 2.0 $3,351,948 E.6.2.2 EXTERNAL FLOODING EVALUATION This SAMA can have an impact on any SBO scenario as well as any seal LOCA scenario. For the external flooding cases, the three flood regimes are impacted differently:
Environmental Report Appendix E  SAMA ANALYSIS Page E-98 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Floods over 310' msl:  In these scenarios, all safety equipment is flooded and the EFW system would not be available. Installation of the damage resistant RCP seals, which is part of this SAMA, would preclude the need to align the primary side makeup/seal injection pump. This would reduce the operator workload slightly improve the reliability of the flood mitigation actions, but the existing HEP is considered to be representative of the difficult set of actions that remain to align secondary side cooling and no reduction of the extreme flood CDF is assumed to occur based on the installation of the N-9000 seals.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (the SBO EDG is flooded) or because a flood warning is not provided and no preparations are made for the flood (the SBO EDG is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of all AC power leads to core damage. These SBO cases are assumed to be completely mitigated by this SAMA
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. As a result, these flooding sequences would be impacted in the same way as the internal events LOOP events. In order to simplify the calculations, SAMA 2 is assumed to eliminate all risk from this flooding sequence. Given the low contribution of these sequences relative to the entire flooding contribution, the impact of this conservative assumption is minimal. The following tables summarize the results of these changes: SAMA 2 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 7.72E-05 166.08 $508,082 Percent Change
-4.8% -6.3% -6.3%  A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-99 License Renewal Application SAMA 2 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 0.00E+00 8.65E-08 0.00E+00 7.72E-05 Base Dose-
 
Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-
 
Risk 132.75 13.13 0.19 1.89 17.87 0.00 0.25 0.00 166.08 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $0 $778 $0 $508,082 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 2 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $14,598,420 $945,053 E.6.2.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $7,300,000 by the TMI staff (Exelon 2007c). E.6.2.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 2 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $3,351,948 $945,053 $4,297,001 $7,300,000 -$3,002,999 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E  SAMA ANALYSIS Page E-100 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.3 SAMA NUMBER 3:  USE NS CCW AS AN ALTERNATE COOLING SOURCE FOR THE DHR HEAT EXCHANGERS (DH-C-1A/B) For LOCAs requiring heat removal with the RHR system, DHRW and DHCCW failures are large contributors to loss of the primary cooling function. Providing the ability to cross-tie the NSCCW system to the DHR heat exchangers would diversify the plant's heat removal capability and eliminate the failures associated with loss of DHRW or DHCCW flow. The hard piped connections are assumed to be sized to allow enough flow to remove decay heat (not just pump cooling loads) and that each division is provided with a cross-connection to NSCCW. E.6.3.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, the NSCCW system was modeled to supply a backup cooling water source in the event that either of the DHCCW trains fails to provide cooling to the DHR heat exchangers. In the event that the DHCCW system is unavailable to provide cooling water on the shell side of either of the DHR heat exchangers, an operator action is required to restore cooling flow via cross-connecting the NSCCW header with the applicable DHR heat exchanger via a remotely operated MOV from within the MCR. The affected model logic was OR gate LPRG0007 for DHR heat exchanger train A and OR gate LPRG0019 for
 
DHR train B. Specifically, the DHCCW train A system top event HA under gate LPRG0007 was replaced with a new AND gate named LPRG0007-1. The inputs to LPRG0007-1 are system top event HA and a new OR gate named NS-1A. The inputs to gate NS-1A are similar to the inputs under the nominal NSCCW system top event NS, with the addition of an MOV event for DHR heat exchanger DH-C-1A and an HEP event that represents operator failure to restore cooling water flow. Likewise, system top event HB under gate LPRG0019 was replaced with a new AND gate named LPRG0019-1. The inputs to LPRG0019-1 are system top event HB and a new OR gate named NS-1B. The inputs to gate NS-1B are similar to the inputs under the nominal NSCCW system top event NS, with the addition of an MOV event for DHR heat Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-101 License Renewal Application exchanger DH-C-1B and the same HEP event described above for restoration of cooling water flow. In addition, all affected logic described above that is modeled within the logic structure for post-LOOP recovery scenarios was also modified, with gate names appended with the characters "-
 
R". The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 3 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.01E-05 29.97 $105,253 Percent Change
-15.2% -0.3% -0.3% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 3 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.54E-07 1.52E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-084.34E-09 3.09E-07 7.28E-071.26E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.60 8.69 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.01 0.91 4.48 0.77 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,621 $42,256 $3,367 $236 $3 $3 $7 $11 $351 $131 $39 $2,778 $14,706 $2,545 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.21E-10 2.08E-11 7.97E-08 8.08E-09 2.24E-072.51E-097.38E-071.82E-072.78E-061.05E-051.69E-08 2.15E-06 1.78E-082.01E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.02 0.30 0.00 1.00 0.25 6.17 2.80 0.00 0.57 0.00 29.97 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $753 $76 $856 $10 $2,819 $695 $17,458 $2,748 $4 $563 $5 $105,253 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-102 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 3 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,995,414 $276,297 2.0 $552,594 E.6.3.2 EXTERNAL FLOODING EVALUATION This SAMA has a very limited impact on external flooding scenarios. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, flood waters fail the DHR system and the SAMA has zero impact.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by the enhanced cooling capabilities of the DHR system as core damage is caused by failure of the flood gates (safety equipment flooded, SBO) or because a flood warning is not provided and no preparations are made for the flood (safety equipment flooded, SBO). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of all AC power leads to core damage. These conditions will cause a seal LOCA and for the small fraction of the scenarios in which power is recovered, the cross-ties could be used to mitigate certain failures. The impact of this SAMA can be approximated by using the baseline internal events model to determine the percent contribution of the "power recovered" SBO sequences to the total SBO contribution. Then, if it is assumed that the relative distribution of "power recovered" sequences for the "E" flood sequence as the same as for the internal events model, the portion of the flood sequence "E" CDF impacted can be calculated. For this evaluation, it is assumed that implementation of this SAMA will eliminate all SBO "power recovered" risk and that the "power recovered" fraction is the same for flood events as it is for internal events SBOs (likely optimistic for the flood case).
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. For simplicity, the CDF for this sequence is
 
assumed to be completely eliminated. Based on the internal events model, SBO sequences contribute a CDF of 3.25E-06/yr while the power recovered SBO sequences contribute only 2.21E-08/yr. This indicates that the "power Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-103 License Renewal Application recovered" SBO evolutions contribute only 0.7 percent of the SBO CDF (2.21E-08 / 3.25E-06/yr
* 100 = 0.7). For flood sequence "E", the expected CDF reduction would then be 2.56E-08 (7.0E-03
* 3.66E-06 = 2.56E-08). The following tables summarize the results of quantification strategy: SAMA 3 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.08E-05 176.71 $540,710 Percent Change
-0.3% -0.3% -0.3% A further breakdown of this information is provided below according to release category. SAMA 3 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.63E-06 8.65E-08 0 8.08E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.63 0.25 0.00 176.71 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,628 $778 $0 $540,710 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 3 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,501,141 $42,332 E.6.3.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $2,450,000 by the TMI staff (Exelon 2007c).
Environmental Report Appendix E  SAMA ANALYSIS Page E-104 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.3.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 3 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $552,594 $42,332 $594,926 $2,450,000 -$1,855,074 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.4 SAMA NUMBER 4:  PROVIDE ALTERNATE POWER TO HPI PUMP MINIMUM FLOW RECIRCULATION VALVES MU-V-36 AND MU-V-37 The current PRA model logic correctly assumes isolation of HPI minimum flow recirculation valves MU-V-36 and 37 on an ESAS, but it does not include the AC power dependences for the "close" action. However, the logic related to opening the minimum flow valves does include the power dependencies, which can result in the generation of cutsets that include the failure to open a flow path that was never isolated. This is critical for the HPI pumps in cases where the HPI flow to the RCS is very low due to the small size of the RCS break/leak. Based on system review and discussions with plant personnel, the only events that could cause the MU-V-36 or MU-V-37 valves to be "stranded closed" are those in which an ESAS based closure occurs when power is available to one or both valves and then one or both of the divisions of valve power fails before the valve(s) can be re-opened to support HPI minimum flow recirculation. A quantification of the contribution from scenarios of this type would require a dynamic PRA model, which is not available to TMI-1. Ho wever, an approximation can be performed to show that risk associated with the MU-V-36/37 design is low and that no SAMAs are required to modify the power supplies to the valves. The current model assumes that power is always available to isolate MU-V-36/37 and if this assumption is accepted for this evaluation, a time weighted probability can be used for power failures to the valves that will approximate the CDF related to "stranding" them closed. In this case, once an ESAS has registered and the HPI pumps are running, 45 minutes are assumed Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-105 License Renewal Application to be available for establishing the minimum flow path before pump failure occurs. For "valve stranding" to be an issue, the loss of power to the valve would have to occur between the time of the ESAS and the time to pump failure. Power failures before the ESAS would not present a problem for minimum flow recirculation because MU-V-36/37 fail "as-is". Power failures after pump failure are not a concern because the pum p will already have failed. Therefore, the pertinent portion of the valve power failure probability is for only 0.75 hours out of 24. Assuming that the likelihood of failure is constant over the 24 hour mission time, this correlates to a fraction of only 3.12E-02. If this fraction is applied to the power inputs for the minimum flow recirculation valve failure logic, a more representative base case will be established with respect to CDF. From this model configuration, the importance of the power supplies for the minimum flow recirculation valves can then be calculated. As mentioned above, this approximation method assumes that power is initially available to isolate the MU-V-36/37 valves, which will not always be the case
 
and overestimates the importance of the power failures. The following table summarizes the changes that were made to the PRA model to establish the new "baseline" used to calculate the importance of the MU-V-36/37 power supply gates: SAMA 4 - Model Changes Gate and / or Basic Event ID and Description Description of Change MRG0001 (existing gate): MR (makeup pump recirculation path) The following inputs were removed from this gate:
* ED1AESV (existing gate): 480V MCC 1A ESV FAILS
* EE1BESV (existing gate): 480V MCC 1B ESV FAILS The following inputs were added to this gate:
* Gate MCC1A-FRACTION
* Gate MCC1B-FRACTION MCC1A-FRACTION (new AND gate) The following inputs were included:
* ED1AESV (existing gate): 480V MCC 1A ESV FAILS
* RECIRC-FRACTION (new basic event: FRACTION OF TIME THAT FAILURE IS CRITICAL FOR MIN FLOW RECIRC Environmental Report Appendix E  SAMA ANALYSIS Page E-106 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 4 - Model Changes Gate and / or Basic Event ID and Description Description of Change RECIRC-FRACTION: FRACTION OF TIME THAT FAILURE IS CRITICAL FOR MIN FLOW RECIRC New basic event representing the fraction of time that a failure of power to the MU-V-36 or 37 valves would result in a "Stranded" valve given that the valve has already closed (3.12E-02). MCC1B-FRACTION (new AND gate) The following inputs were included:
* EE1BESV (existing gate): 480V MCC 1B ESV FAILS
* RECIRC-FRACTION (new basic event: FRACTION OF TIME THAT FAILURE IS CRITICAL FOR MIN FLOW RECIRC Similar changes were made to the LOOP recovered set of logic. The LOOP recovered logic is a reproduction of the base logic without power dependences that is used after power is recovered in a LOOP sequence.
In this case, the RRW value for RECIRC-FRACTION, which captures the importance of both power supplied to the MU-V-36 and 37 valves for both the base and "power recovered" logic, is only 1.006 based on CDF and 1.002 for the Level 2 results, which is below the SAMA screening criteria of 1.01 and demonstrates that changes to the MU-V-36/37 power supply configuration would not be cost beneficial. E.6.5 SAMA NUMBER 5:  ENHANCE VALVES MU-V-76A/B AND MU-V-77A/B TO ALLOW FOR RAPID ALIGNMENT CHANGES IN ACCIDENT CONDITIONS The current MU-V-76A/B and MU-V-77A/B valve configurations do not allow for rapid re-alignment during accident conditions. These valves are used to manipulate the flowpath for the "B" HPI pump between the seal injection and makeup flowpaths, but they also inherently determine whether the "A" or "C" pump can be aligned to the seal injection flowpath. For TMI-1, the capability to quickly align the "C" HPI pump for seal injection would reduce the risk of prominent accident sequences in which thermal barrier cooling has failed in conjunction with the "A" and "B" HPI pumps. Replacing MU-V-76A/B and MU-V-77A/B with MOVs operable from the main control room would allow TMI-1 to use the "C" HPI pump for seal injection and prevent seal LOCAs when the normal cooling methods are unavailable.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-107 License Renewal Application The normal conditions of the plant, which are reflected in the PRA model, show that the "C" pump is the important pump for establishing alternate seal injection and that the benefit is derived from changes to the MU-V-76A/B. However, plant operating practices can change and alterations to the normal alignment of the HPI system could shift the importance to the "A" division. In order to address alternate plant configurations and to provide maximum flexibility, both sets of valves are assumed to require modification. E.6.5.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, cutset changes were made to address the impact of replacing the MU-V-76A/B and MU-V-77A/B valves with MOVs. This method was chosen given that the valve alignment capability can easily be modified through the manipulation of an existing human failure event. In the TMI-1 model, the relevant basic event is the independent HEP INHINJ2_MUHHMUOA, which is set to 1.0 in the baseline model to reflect the inability to locally manipulate the valve in time to support seal injection. In this case, providing the capability to remotely operate the valve is considered to reduce the failure probability to at least 1.0E-02, which is reflected in the cutsets by changing the failure probability of the independent HEP from 1.0 to 1.00E-02. This action is present in a large number of cutsets with multiple other HEPs. Typically, these cases are reviewed as part of the HRA dependency analysis, but for this case, the base probability is 1.0 and the action is not included in any JHEPs because the action always fails due to timing concerns. Setting the probability to something other than 1.0 would normally require inclusion of the action in the dependency analysis to limit the credit
 
taken when dependent conditions exist. No dependency analysis was performed for this SAMA quantification. In this case, excluding the dependency analysis maximizes the benefit of the SAMA and is conservative relative to the identification of cost beneficial SAMAs. The following table summarizes the model changes that were made:
Environmental Report Appendix E  SAMA ANALYSIS Page E-108 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 5 - Model Changes Gate and / or Basic Event ID and Description Description of Change INHINJ2_MUHHMUOA: O PERATOR OPENS CROSS CONNECT VALVES MU-V-76A/B AND STARTS MU-P-1C The basic event probability was changed from 1.0 to 1.00E-02. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 5 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.22E-05 31.49 $109,455 Percent Change
-6.3% -3.4% -2.5% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 5 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 4.86E-114.86E-111.90E-102.46E-103.31E-081.41E-088.34E-09 3.16E-07 6.57E-071.63E-071.69E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.10 0.04 0.02 0.93 4.04 1.00 0.05 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $2 $2 $7 $9 $298 $127 $75 $2,841 $13,271 $3,293 $160 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 7.99E-08 1.43E-08 1.75E-072.75E-097.43E-072.89E-073.12E-061.20E-051.69E-08 2.33E-06 1.91E-082.22E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.24 0.00 1.00 0.39 6.93 3.19 0.00 0.62 0.01 31.49 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $755 $135 $669 $11 $2,838 $1,104 $19,594 $3,134 $4 $610 $5 $109,455 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-109 License Renewal Application SAMA 5 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,157,717 $113,994 2.0 $227,988 E.6.5.2 EXTERNAL FLOODING EVALUATION This SAMA has a very limited impact on external flooding scenarios. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, the MU-V-76A/B and MU-V-77A/B are not used and the SAMA has zero impact.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by the enhanced capabilities of the MU-V-76A/B and MU-V-77A/B valves as core damage is caused by failure of the flood gates (SBO case) or because a flood warning is not provided and no preparations are made for the flood (SBO case). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. These cases will cause a seal LOCA, which is the event this SAMA is primarily designed to prevent. As power recovery could not be performed rapidly enough for SAMA 5 to restore seal cooling and prevent the seal LOCA, the impact of this SAMA on sequence E sequence is negligible and is it is assumed to have no impact on the CDF.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. The CDF for this sequence is assumed to be reduced by the same fraction as the internal events CDF. The following tables summarize the results of quantification strategy: SAMA 5 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.10E-05 177.14 $542,081 Percent Change 0.0% 0.0% 0.0%
Environmental Report Appendix E  SAMA ANALYSIS Page E-110 Three Mile Island Nuclear Station Unit 1 License Renewal Application A further breakdown of this information is provided below according to release category. SAMA 5 - External Flooding Contributions by Release Ca Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.34E-07 8.10E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.35 177.14 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,141 $542,081 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 5 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,541,298 $2,175 E.6.5.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $3,150,000 by the TMI staff (Exelon 2007c). E.6.5.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 5 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $227,988 $2,175 $230,163 $3,150,000 -$2,919,837 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-111 License Renewal Application E.6.6 SAMA NUMBER 6:  ADD CROSS-TIES WITHIN THE TRAINS OF COOLING SYSTEMS - DHR, DHCCW, DHRW Some failure combinations that eliminate both trains of the DHR related cooling systems could be mitigated if cross-ties were available between trains of the DHR, DHRW, and DHCCW systems (not between the systems). For example, these cross-ties would be helpful in conditions where the flow path fails in one train while a pump failure or maintenance event disables the opposite train. To ensure the DHR cross-ties can be implemented in a timely manner for LPI requirements, the associated valves should be operable from the main control
 
room. The use of MOVs in the DHR cross-tie line is beneficial due to the relatively rapid response time required to support low pressure injection; therefore, the MOVs are suggested as part of the design. For the DHCCW and DHRW systems, which support the containment heat removal function of DHR, the time available to respond is much longer. Manual valves could be used for these cross-tie lines and the cross-tie reliability would not be greatly impacted. E.6.6.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. Cross-tie capability for the DHRW system was modeled by adding an AND gate under the gate HAG0001 for train A that was labeled HAG0001-1 and has top event RA (DHRW train A) and gate HAG0001-2 as its inputs. HAG0001-2 is an OR gate that accounts for failure of a proposed crosstie MOV (SAMA6XTMOV1-VAFD), operator failure to perform the crosstie operation (SAMA6-XTIE-HVAOA), failure of a proposed AC power dependency (top event MC, which represents MCC 1C ESV), and the top event for DHRW train B (RB). Similar logic changes were also applied to the model for DHRW train B under gate HBG0001. For DHCCW, the model logic changes for crosstie capability between trains A and B were applied to gates that affected cooling support dependencies for the reactor building spray pumps, the makeup pumps, and DHR pumps. The affected gates for train A systems were Environmental Report Appendix E  SAMA ANALYSIS Page E-112 Three Mile Island Nuclear Station Unit 1 License Renewal Application CSG0018 (reactor building spray pump BS-P-1A), HPGPUMPACOOLSUP1 (makeup pump MU-P-1A), and LPRG0007 (decay heat pump DH-P-1A). The crosstie logic for DHCCW train A, with train B being used as the backup source, is contained under the AND gate HPGPUMPACOOLSUP1-1. This AND gate contains system top event HA and OR gate HPGPUMPACOOLSUP1-2 as its inputs. HPGPUMPACOOLSUP1-2 contains system top HB, the common operator failure event SAMA6-XTIE-HVAOA, gate MC for AC power dependency, and a proposed crosstie MOV (SAMA6XTMOV2-VAFD). Logic for DHCCW train B was revised in a similar fashion for the following affected gates for ECCS train B components: CSG00017 reactor building spray pump BS-P-1B HQGPUMPCCOOLIN makeup pump MU-P-1C LPRG0019 decay heat pump DH-P-1B The AND gate HQGPUMPCCOOLIN-1 (DHCCW train B and train A as backup fail) was used as the cooling support dependency for these three gates identified for train B ECCS components.
The inputs to HQGPUMPCCOOLIN-1 are system top HB (DHCCW train B) and the OR gate HQGPUMPCCOOLIN-2 (DHCCW train A fails as backup). The inputs to gate HQGPUMPCCOOLIN-2 are system top HA, the common operator failure event SAMA6-XTIE-HVAOA, gate MC for AC power dependency, and the proposed crosstie MOV (SAMA6XTMOV2-VAFD). For the DHR system, two system top events representing different functions of this system were affected, namely gate LPI (LPI trains A and B fail), and gate DHR (DHR trains A and B fail). LPI is an AND gate with two inputs:  AND gate LPIA-1 and AND gate LPIB-1. Inputs to LPIA-1 include OR gate LPIA for failure of LPI train A and OR gate LPIA-2, which represents failure of LPI train B to backup train A. LPIA-2 contains the operator failure to perform cross-tie operations (SAMA6-XTIE-HVAOA), power dependency gate MC, gate LPIB for failure of LPI train B, and crosstie MOV failure event SAMA6XTMOV3-VAFD. Likewise, AND gate LPIB-1 contains OR gate LPIB for failure of LPI train B and OR gate LPIB-2, which represents failure of LPI train A to backup train B. LPIB-2 contains the operator failure to perform cross-tie operations (SAMA6-XTIE-HVAOA), power dependency gate MC, gate LPIA for failure of LPI train A, and crosstie MOV failure event SAMA6XTMOV3-VAFD. Identical logic changes were made to system top DHR, which involved AND gate DHRA-1 and AND gate DHRB-1. The only difference is that system top DHRA was used in place of LPIA and DHRB was used in place of LPIB. Similarly, DHRA-1 and DHRA-2 were used in place of LPIA-1 and LPIA-2; and DHRB-1 and DHRB-2 were used in place of LPIB-1 and LPIB-2.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-113 License Renewal Application In addition, all affected logic described above that is modeled within the logic structure for post-LOOP recovery scenarios was also modified, with gate names appended with the characters "-
R". The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 6 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.06E-05 31.00 $108,864 Percent Change
-13.1% -4.9% -3.0% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 6 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 4.86E-114.86E-111.90E-102.46E-103.57E-081.42E-087.69E-09 3.19E-07 6.39E-071.57E-071.60E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.10 0.04 0.02 0.93 3.93 0.97 0.04 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $2 $2 $7 $9 $321 $128 $69 $2,868 $12,908 $3,171 $151 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.40E-08 1.42E-08 1.96E-072.75E-097.81E-072.62E-073.14E-061.03E-051.69E-08 2.33E-06 1.91E-082.06E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.24 0.04 0.26 0.00 1.05 0.35 6.97 2.76 0.00 0.62 0.01 31.00 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $794 $134 $749 $11 $2,983 $1,001 $19,719 $2,705 $4 $610 $5 $108,864 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-114 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 6 Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-Flood AvertedCost-Risk $3,271,711 $3,093,415 $178,296 2.0 $356,592 E.6.6.2 EXTERNAL FLOODING EVALUATION This SAMA has a very limited impact on external flooding scenarios. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, flood waters fail the DHR system and the SAMA has zero impact.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by the addition of the DHR system cross-ties as core damage is caused by failure of the flood gates (safety equipment flooded, SBO) or because a flood warning is not provided and no preparations are made for the flood (safety equipment flooded, SBO). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. These cases will cause a seal LOCA and for the small fraction of the scenario in which power is recovered, the cross-ties could be used to mitigate certain failures. The impact of this SAMA can be approximated by using the baseline internal events model to determine the percent contribution of the "power recovered" SBO sequences to the total SBO contribution. Then, if it is assumed that the relative distribution of "power recovered" sequences for the "E" flood sequence as the same as for the internal events model, the portion of the flood sequence "E" CDF impacted can be calculated. For this evaluation, it is assumed that SAMA implementation will eliminate all SBO "power recovered" risk.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. For simplicity, the CDF for this sequence is
 
assumed to be completely eliminated. Based on the internal events model, SBO sequences contribute a CDF of 3.25E-06/yr while the power recovered SBO sequences contribute only 2.21E-08/yr. This indicates that the "power recovered" SBO evolutions contribute only 0.7 percent of the SBO CDF (2.21E-08 / 3.25E-06/yr
* 100 = 0.7). For flood sequence "E", the expected CDF reduction would then be 2.56E-08 (7.0E-03
* 3.66E-06 = 2.56E-08).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-115 License Renewal Application The following tables summarize the results of quantification strategy: SAMA 6 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.08E-05 176.71 $540,710 Percent Change
-0.3% -0.3% -0.3% A further breakdown of this information is provided below according to release category. SAMA 6 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.63E-06 8.65E-08 0 8.08E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.63 0.25 0.00 176.71 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,628 $778 $0 $540,710 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 6 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,501,141 $42,332 E.6.6.3 COST OF IMPLEMENTATION The cost of installing the powered DHR cross-tie was estimated to be $2,750,000 by the TMI staff (Exelon 2007c). The cross-ties for the DHCCW and DHRW systems are not required to be MOVs due to the longer times available for performing the cross-tie and while there would be a substantial additional cost related to the addition of these cross-ties, only the DHR cross-tie cost of $2,750,000 is used here based on the availability of information.
Environmental Report Appendix E  SAMA ANALYSIS Page E-116 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.6.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 6 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $356,592 $42,332 $398,924 $2,750,000 -$2,351,076 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.7 SAMA NUMBER 7:  USE FIRE SERVICE WATER AS AN ALTERNATE COOLING SOURCE FOR THE ICCW HEAT EXCHANGERS For cases in which NSRW is unavailable due to har dware failures (e.g., flow diversion), the Fire Service Water system could be used to directly cool the ICCW heat exchangers for thermal barrier cooling support. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an
 
action. E.6.7.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. For this model revision, the fire service water system was used to provide a backup cooling water source in the event that NSRW is unavailable to supply cooling water to the ICCW heat exchangers, which in turn renders thermal barrier cooling for the RCP seals unavailable. A new input was added to existing gate SEG0005, which was an AND gate labeled SEG0005-1.
Inputs to this gate included the top event for unavailability of the NSRW system (top event NR) and OR gate SEG0005-2. Inputs to gate SEG0005-2 include the top event for unavailability of Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-117 License Renewal Application the fire service water system (top event FS), a basic event representing mechanical failures associated with this alternate alignment (SAMA7-MECHANICAL), and a HEP event (SAMA7-FSW-HVHOA), which was assigned an assumed failure probability of 0.1 since actions are performed outside the MCR. As a simplification, the failure probability for SAMA7-MECHANICAL was assigned an assumed unavailability of 1.0E-3. Model logic changes were not required for post-LOOP recovery scenarios as seal cooling is not applicable to those accident scenarios. Similar model changes were performed under gate SEG0004 to credit this SAMA for ICCW "B" train cooling. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 7 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.07E-05 30.62 $107,565 Percent Change
-12.7% -6.1% -4.2% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 7 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 4.86E-114.86E-111.90E-102.46E-103.37E-081.41E-088.34E-09 3.15E-07 6.13E-071.52E-071.65E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.10 0.04 0.02 0.92 3.77 0.93 0.05 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $2 $2 $7 $9 $303 $127 $75 $2,832 $12,383 $3,070 $156
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-118 Three Mile Island Nuclear Station Unit 1 License Renewal Application Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 7.99E-08 1.43E-08 1.70E-072.75E-097.43E-072.85E-073.06E-061.06E-051.69E-08 2.31E-06 1.91E-082.07E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.23 0.00 1.00 0.38 6.79 2.83 0.00 0.62 0.01 30.62 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $755 $135 $649 $11 $2,838 $1,089 $19,217 $2,778 $4 $605 $5 $107,565 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 7 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,064,992 $206,719 2.0 $413,438 E.6.7.2 EXTERNAL FLOODING EVALUATION This SAMA can potentially impact scenarios in which AC power is available and the safety equipment has not been flooded. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, all safety equipment is flooded and this SAMA has no impact on the risk.
* Floods between 305' and 310' msl:  Most of the sequences could not be impacted by this SAMA as core damage is caused by failure of the flood gates (safety equipment is flooded) or because a flood warning is not provided and no preparations are made for the flood (safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. In these cases, the ensuing SBO results in a seal LOCA, which is the event SAMA 7 was designed to prevent when power is available. Given that a seal LOCA will occur for sequence "E" whether or not SAMA 7 is implemented, it has no impact on the sequence "E" CDF.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. In order to simplify the quantification of this SAMA, it is assumed that the SAMA 7 eliminates all risk from these floods.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-119 License Renewal Application The following tables summarize the results of these changes: SAMA 7 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.08E-05 176.79 $540,940 Percent Change
-0.3% -0.2% -0.2% A further breakdown of this information is provided below according to release category. SAMA 7 External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 0.00E+00 8.08E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.00 176.79 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $0 $540,940 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 7 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,507,657 $35,816 E.6.7.3 COST OF IMPLEMENTATION Palisades estimated $2.9 million for Fire water cooling to CCW HXs (NMC 2005), Calvert Cliffs estimated $565k for alt DHR cooling (BGE 1998), and Brown's Ferry estimated $1 million for Fire Water to DHR HXs (TVA 2003). The Brown's Ferry estimate is used for TMI. E.6.7.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-120 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 7 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $413,438 $35,816 $449,254 $1,000,000 -$550,746 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.8 SAMA NUMBER 8:  AUTOMATE REACTOR COOLANT PUMP TRIP ON HIGH MOTOR BEARING COOLING TEMPERATURE Seal LOCAs resulting from operator failures to trip the RCPs on loss of motor bearing cooling could be reduced if high temperature sensors were installed on motor bearing cooling water lines to provide automatic trip signals. E.6.8.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate the improved capability of tripping the RCPs upon loss of NSCCW cooling to the motor and pump bearings, the HEP event OTHOT1_RCPTHP1OA was reduced by a factor of 10, from a failure probability of 1.44E-2 to 1.44E-3. Also, to account for the automation of the RCP trip function, all JHEPs including OTHOT1_RCPTHP1OA  were set to 0.0. While the installation of additional trip logic would introduce a previously non-existing source of spurious RCP trip signals that would increase plant risk, no reliable means of estimating the increase in the RCP trip frequency has been identified. As a result, no strategy to quantify the potential increase in risk related to implementation of this SAMA was developed for this quantification. No requantification of the PRA model was required given that all of the changes outlined above could be performed in the cutset files.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-121 License Renewal Application The following table summarizes the data changes that were made: SAMA Number 8 Model Changes Gate and / or Basic Event ID and Description Description of Change OTHOT1_RCPTHP1OA: OPER ATOR FAILS TO TRIP REACTOR COOLANT PUMP ON LOSS OF NSCCW The basic event probability was changed from 1.44E-02 to 1.44E-03. JHHEML-HOT1HEPOA: NSHEML_HER-HP2OA AND OTHOT1_RCPTHP1OA This basic event probability was set to
 
==0.0. JHHNS10HOT1HEPOA==
NS HNS6-----HHXOA AND OTHOT1_RCPTHP1OA This basic event probability was set to
 
==0.0. JHHNS6-HOT1HEPOA==
NS HNS6-----HHXOA AND OTHOT1_RCPTHP1OA This basic event probability was set to
 
==0.0. JHHNSHOTHEOHEPOA==
JHHNS10HOT1HEPOA AND GSHEO1A----HDGOA This basic event probability was set to
 
==0.0. JHHNSHOTHMRHEPOA==
JHHNS10HOT1HEPOA AND MRHMR1-----HMUOA This basic event probability was set to
 
==0.0. JHHOT1-HMR1HEPOA==
OTHOT1_RCPTHP1OA AND MRHMR1-----HMUOA This basic event probability was set to
 
==0.0. JHHOT1-XTIEHEPOA==
OTHOT1_RCPTHP1OA AND NR-NRSRXTIEHVAOA This basic event probability was set to
 
==0.0. JHHOTHMRXTIHEPOA==
OTHOT1_RCPTHP1OA; MRHMR1-----HMUOA; NR-NRSRXTIEHVAOA This basic event probability was set to
 
0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 8 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.06E-5 25.28 $91,111 Percent Change
-13.2% -22.5% -18.8% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 8 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.78E-081.46E-088.54E-09 3.16E-07 6.00E-071.60E-072.05E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Environmental Report Appendix E  SAMA ANALYSIS Page E-122 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 8 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 3.69 0.98 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $340 $131 $77 $2,841 $12,120 $3,232 $194 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.07E-072.75E-097.45E-072.89E-073.06E-071.32E-051.69E-08 2.32E-06 1.91E-082.06E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.28 0.00 1.01 0.39 0.68 3.52 0.00 0.62 0.01 25.28 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $791 $11 $2,846 $1,104 $1,922 $3,459 $4 $608 $5 $91,111 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 8 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,654,373 $617,338 2.0 $1,234,676 E.6.8.2 EXTERNAL FLOODING EVALUATION This SAMA has no impact on external flooding scenarios. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, flood waters fail the safety equipment and the SAMA has zero impact.
* Floods between 305' and 310' msl:  Most of the sequences could not be impacted by this SAMA as core damage is caused by failure of the flood gates (safety equipment is flooded) or because a flood warning is not provided and no preparations are made for the flood (safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. In these cases the LOOP trips the RCPs so the auto trip function is not required. In addition, the ensuing SBO results in a seal LOCA, which is the event SAMA 8 was designed to prevent. Given that a Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-123 License Renewal Application seal LOCA will occur for sequence "E" whether or not SAMA 8 is implemented, it has no impact on the sequence "E" CDF.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. Given that a LOOP event will trip the RCPs, SAMA 8's auto trip function is not required and it has no impact on these flood sequences. In summary, this SAMA has no measurable impact on the external flooding contributors, as shown in the following tables: SAMA 8 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 8 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 8 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.8.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $145,000 by the TMI staff (Exelon 2007c).
Environmental Report Appendix E  SAMA ANALYSIS Page E-124 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.8.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 8 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $1,234,676 $0 $1,234,676 $145,000 $1,089,676 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive. E.6.9 SAMA NUMBER 9:  PROCEDURALIZE LOCAL ADV OPERATION TMI-1 has procedures to perform the local ADV operations that are not credited in the PRA model (the failure probability is set to 1.0). If the available procedures are credited and used to allow local operation of the ADVS for cooldown/depressurization after loss of remote capability, the RRW value of the operator action would be reduced below the SAMA review threshold.
This SAMA is used demonstrate that the RRW for this operator action would be below the SAMA review threshold if appropriate credit were taken and that no SAMAs are required to address local ADV operations. For this case, an HEP of 0.1 is assumed for the action (AV-LOCADV--HCDOA). The model does not contain any JHEPs that include AV-LOCADV--HCDOA; therefore, no additional changes are required. This change was made directly in the cutsets and no model requantification was required, as summarized below: SAMA 9 - Model Changes Gate and / or Basic Event ID and Description Description of Change AV-LOCADV--HCDOA: OPERATOR ACTION FAILURE TO LOCALLY OPERATE ADVS ON LOSS OF AIR Basic event probability changed from
 
1.0 to 1.00E-01. In this case, the RRW value for AV-LOCADV--HCDOA was reduced to 1.005 for CDF and 1.004 for the Level 2 results. As these are both below the SAMA screening criteria of 1.01, this assessment demonstrates that enhancing local ADV operation would not be cost beneficial.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-125 License Renewal Application E.6.10 SAMA NUMBER 10:  AUTOMATE BWST REFILL Failure to refill the BWST is a large contributor to some SGTR sequences, especially those in which the main steam ADVs fail to operate (including operator errors). Automating the refill function would improve the reliability of this process and reduce the contributions from prominent SGTR sequences by providing a long term high pressure injection source. While isolation of the break is a more desirable approach to mitigating SGTR events, providing long term primary side injection is a potential means of preventing core damage and is considered to result in a success path by providing time to cool down the RCS and to recover isolation capability. Automation of the BWST refill function will require linking tank level sensors/transmitters with logic that will start the transfer pumps, open the valves in the flowpath, and return the system to standby when the tank is refilled. This SAMA also requires that an adequate volume of boron will be available for at least 24 hours (without operator intervention) given the largest expected leak rate for SGTR initiating events. It is possible that refill of the BWST would be capable of mitigating some ISLOCA events, but because an evaluation of Auxiliary Building flooding from ISLOCA flow has not been performed, no credit is taken for ISLOCA cases. E.6.10.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of SAMA 10's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, cutset changes were made to address the impact of automating the BWST refill function. This method was chosen given that BWST refill reliability can easily be modified through the manipulation of existing human failure events. In the TMI-1 model, the relevant basic events include an independent event as well as joint human error events. In this case, automating operation of the refill system (with human backup) is considered to reduce the failure probability to at least 1.0E-04, which is reflected by changing the failure probability of the independent HEP from 2.65E-02 to 1.0E-04. Because automation Environmental Report Appendix E  SAMA ANALYSIS Page E-126 Three Mile Island Nuclear Station Unit 1 License Renewal Application of the function basically removes it from the joint human error events, those events are set to 0.0. If the combinations of the remaining actions are important to the model, they would be treated in separate events and the development of new combinations is not required. The following table summarizes the model changes that were made: SAMA 10 - Model Changes Gate and / or Basic Event ID and Description Description of Change BWST-HRE27-HTKOA:  FAILURE TO REFILL BWST (SPLIT FRAC REV) The basic event probability was changed from 2.65E-02 to 1.0E-04. JHAHCD4RE27HEPOA: AVHCD4_FF--HCDOA AND BWST-HRE27-HTKOA (JHEP addressing BWST refill and cooldown via secondary side) The basic event probability was changed from 9.17E-05 to 0.0.
JHHRE27HL1AHEPOA: BWST-HRE27-HTKOA AND DLHHL1A----HVHOA (JHEP addressing BWST refill and opening drop line for DHR cooling) The basic event probability was changed from 2.00E-04 to 0.0. JHHEF2HRE27HEPOA: AVHEF2_FF--HCDOA AND BWST-HRE27-HTKOA (JHEP addressing BWST refill and manually initiating cooldown using the OTSG) The basic event probability was changed from 1.3E-03 to 0.0. JHHCD5HRE27HEPOA: DPHCD5-FF--HDPOA AND BWST-HRE27-HTKOA (JHEP addressing BWST refill and manual pressurization with the RCPs unavailable) The basic event probability was changed from 1.90E-04 to 0.0. JHHIGHREHHLHEPOA: IGHIG1_HER-HSGOA, BWST-HRE27-HTKOA, and DLHHL1A----HVHOA (JHEP addressing BWST refill, failure to isolate a SGTR, and opening drop line for DHR cooling) The basic event probability was changed from 5.0E-07 to 0.0. (Event was not in cutsets) The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 10 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.29E-05 28.06 $90,062 Percent Change
-3.4% -14.0% -19.8% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-127 License Renewal Application SAMA 10 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 5.86E-08 1.19E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 0.34 6.81 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $1,629 $33,082 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.29E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 28.06 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $90,062 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 10 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,780,687 $491,024 2.0 $928,048 E.6.10.2 EXTERNAL FLOODING EVALUATION This SAMA is of importance in SGTR events where RCS inventory leaves the containment and is unavailable for recirculation from the sump. For the external flooding cases, this is not an issue as the reactor is tripped by a manual shutdown rather than an SGTR event. While LOCAs are likely in external flooding scenarios due to SBO induced seal LOCAs, the sump would be available if AC power was subsequently recovered. No measurable risk reduction is believed to result from implementation of this SAMA for external flooding, as shown below:
Environmental Report Appendix E  SAMA ANALYSIS Page E-128 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 10 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 10 - External Flooding Contributions by Release Ca Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood FrequencyBase Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 1.33E+02 1.31E+01 1.90E-01 1.89E+00 1.79E+01 1.07E+01 2.50E-01 3.70E-01 1.77E+02 SAMA Dose-Risk 1.33E+02 1.31E+01 1.90E-01 1.89E+00 1.79E+01 1.07E+01 2.50E-01 3.70E-01 1.77E+02 Base OECR 4.06E+05 4.01E+04 5.98E+02 5.77E+03 5.48E+04 3.29E+04 7.78E+02 1.22E+03 $542,159 SAMA OECR 4.06E+05 4.01E+04 5.98E+02 5.77E+03 5.48E+04 3.29E+04 7.78E+02 1.22E+03 $542,159 The external flooding component of the averted cost-risk for this SAMA is, therefore, $0: SAMA 10 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.10.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $3,800,000 by the TMI staff (Exelon 2007c). E.6.10.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 10 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $982,048 $0 $982,048 $3,800,000 -$2,817,952 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-129 License Renewal Application Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.11 SAMA NUMBER 11:  ENHA NCE EXTREME EXTERNAL FLOODING MITIGATION EQUIPMENT TO ADDRESS SBO AND LOSS OF SEAL COOLING SCENARIOS Making the extreme flooding equipment proposed in SAMA 32 useful for SBO conditions, especially those with TD EFW failure, would require permanently mounting the submersible pumps so that the suctions could easily be swapped from a piped water source to the flood water source. Permanently installing the portable generator and the pumps so that they could be auto aligned (and manually aligned from the MCR should auto alignment fail) to support seal cooling would address both SBO and non-SBO loss of seal cooling cases through the ability to rapidly align alternate seal cooling. It is recognized that the requirements of this SAMA are extreme, but in order to mitigate an SBO with EFW failures, it is necessary to provide alternate power to support a means of heat removal. Long term heat removal can be accomplished either by maintaining primary integrity (through RCP seal protection) and using the secondary side systems for heat removal, or through some form of a feed and bleed method. However, a feed and bleed method requires a DHR system that will allow recirculation in order to prevent containment overfill. The added complexity of installing an SBO capable DHR system is considered to be at least as difficult as automating the 480V AC generator alignment, which is proposed by this SAMA. While this SAMA has been retained on the SAMA list due to flooding considerations, the simpler solution to providing long term SBO survivability given EFW failure for internal event initiators is considered in SAMA 24. E.6.11.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA.
Environmental Report Appendix E  SAMA ANALYSIS Page E-130 Three Mile Island Nuclear Station Unit 1 License Renewal Application To simulate implementation of this SAMA, the cutsets from SAMA 1 were used as a starting point as they addressed the ability to prevent a seal LOCA given failure of the "A" and "B" EDGs. In order to capture the additional SAMA 11 capabilities of providi ng core cooling in an SBO even with turbine driven EFW failure, the important EDG and AFW equipment failures were set to zero. Setting these events to zero simulates recovery from these failures by the SAMA 11 equipment. The following table lists the basic event data changes that were made to the SAMA 1 cutset file to quantify the impact of this SAMA: SAMA 11 - Model Changes Gate and / or Basic Event ID and Description Description of Change EFEF-P-1----P7FS:  TURBINE-DRIVEN PUMP EF-P-1 FAILS TO START The basic event probability was changed from 4.66E-03 to 0.0.
EFEFP1------P7FR: TURBINE-DRIVEN PUMP EF-P-1 FAILS TO RUN The basic event probability was changed from 5.06E-02 to 0.0. EF-CCFEFW-LETHAL: LETHAL SHOCK TO THE EFW SYSTEM DUE TO COMMON CAUSE FAILURES The basic event probability was changed from 4.25E-04 to 0.0.
GA1ADG------DGFS: DIESEL GENERATOR 1A FAILS TO START The basic event probability was changed from 1.13E-02 to 0.0. GA-EDG-1A---DGFR: DIESEL 1A FAILS TO RUN The basic event probabilit y was changed from 2.07E-02 to 0.0. GA-EG-Y-1A--DGMM: Emergency Diesel Generator 1A in Maintenance The basic event probability was changed from 1.61E-02 to 0.0.
GB1BDG------DGFS: DIESEL GENERATOR 1B FAILS TO START The basic event probability was changed from 1.13E-02 to 0.0. GB-EDG-1B---DGFR: DIESEL 1B FAILS TO RUN The basic event probabilit y was changed from 2.07E-02 to 0.0. GB-EG-Y-1B--DGMM: Emergency Diesel Generator 1B in Maintenance The basic event probability was changed from 1.61E-02 to 0.0.
GSEG-Y-4----DGFS: STATIO N BLACKOUT DG FAILS TO START The basic event probability was changed from 1.13E-02 to 0.0. GS-SBODG----DGFR: SBO DIESEL FAILS TO RUN The basic event probabilit y was changed from 2.07E-02 to 0.0. GS-EG-Y-4---DGMM: SBO Diesel Generator in Maintenance The basic event probability was changed from 1.30E-2 to 0.0. GA-1A1BSBO-CDGFR: EDG CCF Run DG-1A;DG-1B;DG-SBO The basic event probability was changed from 1.53E-04 to 0.0.
GAEDG-STARTCDGFS: EDG Fa il to Start CCF DG-All 3 The basic event probability was changed from 5.25E-05 to 0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-131 License Renewal Application SAMA 11 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 1.57E-05 24.43 $87,640 Percent Change
-33.8% -25.1% -21.9% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 11 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.47E-07 1.22E-06 1.80E-07 1.16E-08 4.21E-114.21E-112.37E-116.68E-111.11E-081.10E-083.35E-10 2.28E-07 5.18E-077.85E-071.22E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.56 6.98 0.91 0.06 0.00 0.00 0.00 0.00 0.03 0.03 0.00 0.67 3.19 4.83 0.03 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,427 $33,916 $3,348 $216 $2 $2 $1 $3 $100 $99 $3 $2,050 $10,464 $15,857 $115 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 3.41E-11 0.00E+00 8.41E-09 8.11E-10 1.24E-074.24E-107.80E-081.32E-088.26E-071.09E-051.25E-08 3.38E-07 4.81E-101.57E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.02 0.00 0.17 0.00 0.11 0.02 1.83 2.90 0.00 0.09 0.00 24.43 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $0 $0 $79 $8 $474 $2 $298 $50 $5,187 $2,849 $3 $89 $0 $87,640 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 11 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,452,029 $819,682 2.0 $1,639,364
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-132 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.11.2 EXTERNAL FLOODING EVALUATION The severe flooding guidelines were originally credited in the IPEEE for both floods above 310' msl as well as for floods between 305' and 310' msl. Due to a more limited preparation time for the 305' to 310' msl floods, the failure probability was assumed to be 0.5 rather than the 0.255 used for the 310' msl floods. For floods below 305' msl, no credit was taken for the severe flooding guidelines as the submersible pumps used for secondary side makeup require flood water in the turbine building for a suction source. Given that this SAMA includes provisions for an alternate secondary side pump suction source, it is assumed that credit could be taken for the floods below 305', as well. The credit taken for this SAMA will be the same for all flood scenarios given that the proposed changes will reduce the manipulation time to a point where it is short (within 13 minutes for auto alignments cause by undervoltage) in comparison to the available time for all of the scenarios (on the order of 18-24 hours from the action cue). This factor reduces the impacts of time stress on the alignment failure probability. For the purposes of this analysis, implementation of this SAMA is assumed to reduce the HEP for alignment of the external flooding measures from 1.1E-01 to 1.0E-04. The large reduction is based on the fact that SAMA 11 automates the system response and no operator action is required. As a result, there is no need to consider operator dependence factors for the initiation failure probability. In addition, the availability of the diverse, alternate portable AC generator is considered to reduce the failure probability of the flood-safe AC power source from 1.43E-01 to 2.04E-02 (1.43E-01
* 1.43E-01 = 2.04E-2, which assumes completely independent generators).
This results in a total failure probability of 2.05E-02 (1.0E-04 + 2.04E-02 = 2.05E-02) for the severe flooding mitigation strategy. Because the severe flooding guidelines were credited differently in each of the flood ranges, three separate strategies are required to obtain the revised core damage frequencies for the flooding scenarios:
* Floods >310' msl:  The CDF for this scenario was calculated in the IPEEE as the product of the flood frequency and the failure probability for the alignment of the severe flooding mitigation strategy. As a result, the revised frequency can be obtained by multiplying the base CDF by the ratio of SAMA based severe flood mitigation failure probability to the baseline severe flood mitigation failure probability (2.05E-02 / 2.55E-01 = 8.03E-02).
* Floods between 305' and 310' msl:  In the IPEEE, a multiplier of 0.5 was applied to each of Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-133 License Renewal Application the sequences in the flooding event tree to represent the potential to avert the flood using the severe flooding guidelines. The CDFs for these sequences can be made to reflect implementation of this SAMA by multiplying each sequence specific CDF by the ratio of SAMA based severe flood mitigation failure probability to the baseline severe flood mitigation failure probability (2.05E-02 / 5.0E-01 = 4.10E-02).
* Floods below 305' msl:  No credit was taken for the severe flooding guidelines for these cases in the IPEEE and as a result, the CDF can be directly multiplied by 2.05E-02. The results of this process are summarized below:  SAMA 11 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 5.81E-06 12.45 $38,036 Percent Change
-92.8% -93.0% -93.0% A further breakdown of this information is provided below according to release category. SAMA 11 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood FrequencyBase Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 5.12E-06 2.53E-07 2.67E-09 3.63E-08 2.45E-07 1.47E-07 3.47E-09 5.13E-09 5.81E-06 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 10.66 0.53 0.01 0.08 0.72 0.43 0.01 0.01 12.45 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $32,598 $1,610 $24 $231 $2,199 $1,318 $31 $25 $38,036 The external flooding based averted cost-risk for this SAMA is shown below: SAMA 11 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $1,094,145 $14,449,328 E.6.11.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $4,250,000 by the TMI staff (Exelon 2007c).
Environmental Report Appendix E  SAMA ANALYSIS Page E-134 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.11.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 11 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $1,639,364 $14,449,328 $16,088,692 $4,250,000 $11,838,692 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive. E.6.12 SAMA NUMBER 12:  USE THE DHR SYSTEM AS AN ALTERNATE SUCTION SOURCE FOR HPI Failures of the BWST suction path (MU-V-14A/B) to the HPI pumps will lead to core damage in
 
scenarios requiring early makeup. Through implementation of procedure changes, the DHR system could be aligned to take suction from the BWST and supply flow to the HPI system to allow injection in these cases. While the events that will cause failure of the HPI suction path are low probability events, the options to prevent core damage in those cases are extremely limited. The existing DHR and HPI piping provide an alternate path that could be used and credited if plant procedures and training were modified. E.6.12.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, the PRA model was changed to accommodate existing logic for valves DH-V-7A/B in the HPI system injection path logic. A new human error Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-135 License Renewal Application probability (HEP) event was added with a screening value of 0.1 (SAMA12-DHMUHVAOA) in conjunction with valve hardware failure events and power dependencies. Logic representing the dependence on the DHR system itself was not included in the DH-V-7A/B suction path. Inclusion of the DHR dependence would reduce the averted cost-risk calculated for this SAMA, but the impact is estimated to be small given that the alternate suction path failures would be dominated by the 0.1 failure probability of the operator action and the common valve power dependences. The changes made to the model are summarized in the following table: SAMA 12 - Model Changes Gate and / or Basic Event ID and Description Description of Change Gate HPG00MAC: NO FLOW FROM PUMP MU-P1A Deleted the following gate:
* Gate HPG00MBK: NO FLOW FROM MU-V-14AOR MU-V-14B Added the following gate:
* Gate HPG00MBK-1 (new): NO SUCTION SOURCE FOR HPI Gate HPG00MBK-1: NO SUCTION SOURCE
 
FOR HPI New AND gate representing the availability of both the BWST and the DHR heat exchangers as injection suction sources. The gate includes the following input:
* Gate HPG00MBK (existing): NO FLOW FROM MU-V-14AOR MU-V-14B
* Gate HPG00MBK-2 (new): HPI SUCTION VIA DH-V-7 MOVS Gate HPG00MBK-2: HPI SUCTION VIA DH-V-7
 
MOVS New OR gate representing the DHR system suction path for HPI. The gate includes the following input:
* Gate HL (existing): HL (DH-V-7A/B failures) Basic event SAMA12-DHMUHVAOA: OPERATOR FAILS TO ALIGN DHR TO MAKEUP PUMP 
 
SUCTION New basic event representi ng the probability that the operators will fail to align the DHR system as the suction injection mode suction source for HPI. 
 
Failure probability = 0.1. Similar changes have been made to the "power recovered" logic. The "power recovered" logic is used in portions of the LOOP tree in which offsite power has been restored and the power dependencies of the logic are removed to preclude failure of OSP from disabling equipment.
Similar changes were also made to credit the MU-P-1B and MU-P-1C pumps with the alternate injection suction alignment from the DHR system.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-136 Three Mile Island Nuclear Station Unit 1 License Renewal Application The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 12 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.27E-05 31.64 $109,292 Percent Change
-4.2% -3.0% -2.6% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 12 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.47E-07 1.56E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 6.80E-071.63E-072.06E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.56 8.92 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.18 1.00 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,427 $43,368 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $13,736 $3,293 $195 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual Risk Freq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.11E-072.75E-097.44E-072.89E-073.14E-061.24E-051.68E-08 2.33E-06 1.91E-082.27E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.28 0.00 1.00 0.39 6.97 3.31 0.00 0.62 0.01 31.64 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $806 $11 $2,842 $1,104 $19,719 $3,251 $4 $610 $5 $109,292 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 12 - Non-External Flooding Averted Cost Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,172,492 $99,219 2.0 $198,438
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-137 License Renewal Application E.6.12.2 EXTERNAL FLOODING EVALUATION This SAMA is of importance in LOCA events where RCS inventory makeup is required using the HPI system suction from the BWST. For the external flooding cases, this is not an issue as the reactor is tripped by a manual shutdown (or a LOOP) rather than a LOCA event. While LOCAs are likely in external flooding scenarios due to SBO induced seal LOCAs, the HPI system would be unavailable in those cases during the SBO. There is some potential for this SAMA to provide a benefit in the flood induced SBO scenarios where AC power is recovered prior to core damage, but the likelihood of recovering power in the short amount of time to prevent core damage in believed to be very low for flood conditions and this SAMA is assumed to provide zero benefit. As a point of reference, the SBO sequences from the base model were quantified and the resulting cutsets were reviewed to determine the contribution of BWST suction failures after power recovery. There was no measurable contribution from suction path failures. The sequences quantified included those in which AC power was both recovered and not recovered, specifically:
* LOOP-055
* LOOP-057
* LOOP-058
* LOOP-059
* LOOP-062
* LOOP-064
* LOOP-066
* LOOP-067
* LOOP-068
* LOOP-069 As a result, this SAMA would not yield a measurable risk reduction for the external flooding events, as shown below:
Environmental Report Appendix E  SAMA ANALYSIS Page E-138 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 12 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 12 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 The external flooding component of the averted cost-risk for this SAMA is, therefore, $0: SAMA 12 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.12.3 COST OF IMPLEMENTATION Procedure changes are estimated to be $50,000 (CPL 2004). E.6.12.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-139 License Renewal Application SAMA 12 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $198,438 $0 $198,438 $50,000 $148,438 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive. E.6.13 SAMA NUMBER 13:  CHANGE IA SYSTEM LOGIC TO AUTOMATICALLY START IA-P-1A/B AFTER A LOW VOLTAGE TRIP IN CONJUNCTION WITH AN ESAS The current IA system logic requires the operators to re-load the IA compressors on emergency power after a low voltage trip when an ESAS is registered. Automating the re-loading of these compressors would remove the requirement for the operators to perform this task in accident conditions. The scenarios of interest for this SAMA are turbine building steam line breaks that cause both a LOOP (due to adverse environmental conditions) and an ESAS, which will require the operators to reload the IA compressors. The importance of automating this action is driven by the short time that is available to prevent loss of seal cooling due to closure of MU-V-20, IC-V-3, and IC-V-4. The PRA indicates that the air supplies for these valves will deplete in 20 minutes after loss of IA and will go closed. While recovery may be possible after the initial closure, no credit is taken for such recovery actions in the model. E.6.13.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. The HEP event for failure to manually start the air compressors using emergency power from the station diesel generators (AMHAM2-----HC1OA) was changed from a failure probability of 8.88E-2 to 1.00E-05 to simulate the improved reliability due to proposed automatic restart logic.
In addition, the JHEPs including AMHAM2-----HC1OA were set to zero to account for the Environmental Report Appendix E  SAMA ANALYSIS Page E-140 Three Mile Island Nuclear Station Unit 1 License Renewal Application removal of the operator from the dependence chain. The following table summarizes the changes that were made to the basic event data: SAMA 13 - Model Changes Gate and / or Basic Event ID and Description Description of Change AMHAM2-----HC1OA:  Basic event probability changed from 8.88E-02 to 1.00E-05. JHHAM2-HEF1HEPOA: AMHAM2-----HC1OA AND EFHEF1_OPERH2HOA Basic event probability set to 0.0. JHHAM2HINJ1HEPOA: AMHAM2-----HC1OA AND INHINJ1_MUHHMUOA Basic event probability set to 0.0. JHHAM2HINJ4HEPOA: AMHAM2-----HC1OA AND INHINJ4_MUHHVCOA Basic event probability set to 0.0. JHHAMHEFHBWHEPOA: JHHAM2-HEF1HEPOA AND BWHBW1-----HP2OA Basic event probability set to 0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 13 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.30E-05 31.17 $106,172 Percent Change
-3.0% -4.4% -5.4% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 13 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.00E-07 1.47E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.89E-081.46E-088.54E-09 3.16E-07 7.00E-071.64E-072.15E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.29 8.41 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.31 1.01 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $11,120 $40,866 $3,367 $236 $3 $3 $7 $11 $350 $131 $77 $2,841 $14,140 $3,313 $203
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-141 License Renewal Application Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.19E-072.75E-097.44E-072.89E-073.16E-061.28E-051.40E-08 2.34E-06 1.91E-082.30E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.00 0.39 7.02 3.41 0.00 0.62 0.01 31.17 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $837 $11 $2,842 $1,104 $19,845 $3,349 $4 $613 $5 $106,172 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 13 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,119,064 $152,647 2.0 $305,294 E.6.13.2 EXTERNAL FLOODING EVALUATION This SAMA will have no measurable benefit for external flooding cases given that equipment is either flooded, an SBO and subsequent seal LOCA occurs (the main goal of this SAMA is to prevent seal LOCAs), or a LOOP will occur without an ESAS signal, as summarized below: Floods over 310' msl:  In these scenarios, all safety equipment is flooded and this SAMA has no impact on the risk. Floods between 305' and 310' msl:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (safety equipment if flooded) or because a flood warning is not provided and no preparations are made for the flood (safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. These scenarios will result in an SBO and a subsequent seal LOCA independent of the implementation status of SAMA 13. Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. Given that IA will not require manual reload without a coincident ESAS and that an ESAS is not expected for a LOOP without a seal LOCA, implementation of SAMA 13 for seal LOCA prevention will not be beneficial. Consequently, this SAMA would not yield a measurable risk reduction for the external flooding events, as shown below:
Environmental Report Appendix E  SAMA ANALYSIS Page E-142 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 13 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 13 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 13 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.13.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $950,000 by the TMI staff (Exelon 2007c). E.6.13.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-143 License Renewal Application SAMA 13 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $305,647 $0 $305,647 $950,000 -$644,706 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.14 SAMA NUMBER 14:  REPLACE HPI PUMP COOLING ALIGNMENT VALVES WITH MOVS In the event that the normally aligned cooling source to a HPI pump fails, the current plant configuration requires local operation of the valves to swap the pump to the alternate cooling source. The time required to perform this action is considered to preclude it as a means of both preventing seal LOCAs in loss of seal cooling evolutions and for providing high pressure makeup. Replacing the valves with MOVs would allow the operators to rapidly align the alternate cooling source from the MCR in time to prevent a seal LOCA or provide high pressure injection. E.6.14.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, exc l udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. The ability to cross-connect cooling systems for the makeup pumps necessitated a change to the model logic for all three makeup pumps depending upon their ESAS alignments. The following paragraphs outline the model changes for each of the makeup pumps: Makeup Pump A Aligned to ESAS Train A:
The system top event HA under AND gate HPGPUMPACOOLSUP1 was replaced with an AND gate named HPGPUMPACOOLSUP1-1, which contained top event HA and an OR gate named HPGPUMPACOOLSUP1-2 as its inputs. The inputs to HPGPUMPACOOLSUP1-2 were a new Environmental Report Appendix E  SAMA ANALYSIS Page E-144 Three Mile Island Nuclear Station Unit 1 License Renewal Application HEP event (SAMA14-HEP-HVAOA), a basic event that accounted for combined mechanical and electrical failures (SAMA14AMECH-ELEC), and system top event NS for unavailability of the NSCCW system. The HEP event was assigned a failure probability of 0.01, which was based on assuming all actions required for realigning cooling water were capable of being performed from inside the main control room. The event SAMA14AMECH-ELEC was estimated to have an unavailability of 0.01, based on a generic combination of mechanical and electrical support dependency failures unique to makeup pump MU-P-1A. Makeup Pump A Aligned for RCP Seal Injection:
The system top event NS under AND gate HPGPUMPACOOLSUP2 was replaced with an AND gate named HPGPUMPACOOLSUP2-1, which contained the system top event NS and a new OR gate named HPGPUMPACOOLSUP2-2. This new OR gate contained the HEP event SAMA14-HEP-HVAOA and basic event SAMA14 AMECH-ELEC, which were both described above, and the system top event HA, simulating the loss of DHCCW train A. Makeup Pump B Cooling Water Dependency:
The physical arrangement of the MU-P-1B cooling piping is such that complex back feeding and the installation of multiple, additional MOVs would be required to allow DHCCW to be used for pump cooling in place of NSRW. Exelon's cost estimate for this SAMA does not include the costs associated with these types of changes; however, credit is taken in this evaluation for alternate MU-P-1B cooling. This conservative approach was used in order to provide a bounding assessment of the benefit related to alternate HPI pump cooling without expending the additional resources that would be required to fully develop the costs of providing DHCCW to MU-P-1B. The system top event NS under OR gate HPGPUMPBCOOL was replaced with an AND gate named HPGPUMPBCOOL-1, which contained the system top event NS and a new OR gate named HPGPUMPBCOOL-2. This new OR gate contained the HEP event SAMA14-HEP-HVAOA described above and a new basic event SAMA14BMECH-ELEC, which was assigned an unavailability of 0.01, based on a generic combination of mechanical and electrical support dependency failures unique to makeup pump MU-P-1B. In addition, HPGPUMPBCOOL-2 also contained the system top event HA, simulating the loss of DHCCW train A.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-145 License Renewal Application Makeup Pump C Aligned to ESAS Train B:
The system top event HB under OR gate HQGPUMPCCOOLIN was replaced with an AND gate named HQGPUMPCCOOLIN-1, which contained top event HB and a new OR gate named HQGPUMPCCOOLIN-2 as its inputs. The inputs to HQGPUMPCCOOLIN-2 were the HEP event SAMA14-HEP-HVAOA described above, a basic event that accounted for combined mechanical and electrical failures unique to makeup pump MU-P-1C (SAMA14CMECH-ELEC), and system top event NS for unavailability of the NSCCW system. In addition, all affected logic described above that is modeled within the logic structure for post-LOOP recovery scenarios was also modified, with gate names appended with the characters "-
 
R". The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 14 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 1.97E-05 29.86 $105,634 Percent Change
-16.9% -8.4% -5.9% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 14 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 4.86E-114.86E-111.90E-102.46E-103.29E-081.42E-087.69E-09 3.19E-07 5.34E-071.57E-071.60E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.10 0.04 0.02 0.93 3.28 0.97 0.04 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $2 $2 $7 $9 $296 $128 $69 $2,868 $10,787 $3,171 $151
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-146 Three Mile Island Nuclear Station Unit 1 License Renewal Application Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.40E-08 1.42E-08 1.63E-072.75E-097.81E-072.62E-073.01E-069.70E-061.69E-08 2.32E-06 1.91E-081.97E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.24 0.04 0.22 0.00 1.05 0.35 6.69 2.59 0.00 0.62 0.01 29.86 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $794 $134 $623 $11 $2,983 $1,001 $18,928 $2,542 $4 $608 $5 $105,634 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 14 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,987,676 $284,035 2.0 $568,070 E.6.14.2 EXTERNAL FLOODING EVALUATION This SAMA will have limited benefit for external flooding cases given that equipment is either flooded, an SBO occurs, or the combined probability of the flood initiators with loss of HPI pump cooling evolutions is so low that the SAMA will not provide a measurable risk reduction, as summarized below: Floods over 310' msl:  In these scenarios, all safety equipment is flooded and this SAMA has no impact on the risk. Floods between 305' and 310' msl:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (safety equipment if flooded) or because a flood warning is not provided and no preparations are made for the flood (safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. These are SBO scenarios in which SAMA 14 would not typically provide a benefit. In the event that AC power is recovered before core damage occurs, SAMA 14 could be beneficial if HPI pump cooling was also lost in the evolution; however, review of the baseline SBO sequence importance list shows that the largest RRW for any DHCCW or DHRW event is 1.001 and all NSCCW events fell below the truncation limit of the quantification and are not even included in the importance list. Therefore, no credit is taken for this SAMA in flood sequence "E".
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-147 License Renewal Application Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. There is some potential for SAMA 14 to provide a benefit for these cases and for the purposes of simplifying the quantification, SAMA 14 is assumed to eliminate all risk for these flood evolutions. The following tables summarize the results of quantification strategy: SAMA 14 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.08E-05 176.79 $540,940 Percent Change
-0.3% -0.2% -0.2% A further breakdown of this information is provided below according to release category. SAMA 14 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 0.00E+00 8.08E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.00 176.79 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $0 $540,940 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 14 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,507,657 $35,816 E.6.14.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $3,150,000 by the TMI staff (Exelon 2007c).
Environmental Report Appendix E  SAMA ANALYSIS Page E-148 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.14.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 14 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $568,070 $35,816 $603,886 $3,150,000 -$2,546,114 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.15 SAMA NUMBER 15:  AUTOMATIC SWAP TO RECIRCULATION MODE The operator action to swap to recirculation mode is a key action for LOCA scenarios.
Automating this function would improve the reliability of this action, especially in the rapidly evolving events where other actions are competing for the attention of the operators. This SAMA should provide the capability to automatically align high or low pressure recirculation mode, depending on the conditions of the plant. E.6.15.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate the automatic swapover from injection to recirculation, the HEP events SAHSR1-----HSROA (for large LOCAs) and SAHSR2-----HSROA (for non-large LOCAs) were set to 1.00E-05 to simulate automation of the action. The corresponding JHEP was set to 0.0 to capture the removal of the recirculation action from the dependence chain. Given that these changes include only the modification of basic event probabilities, the changes were made to the cutset Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-149 License Renewal Application files and no model requantification was required. The cutset changes are summarized in the following table: SAMA 15 - Model Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE SAHSR1-----HSROA: OPERATOR FAIL S TO TAKE PROPER ACTION WITHIN ONE MINUTE The basic event probability was changed from 1.71E-02 to 1.00E-05.
SAHSR2-----HSROA: OPERATOR FAIL S TO TAKE PROPER ACTION WITHIN TEN MINUTE The basic event probability was changed from 1.30E-04 to 1.00E-05. JHHHL1AHSR2HEPOA: DLHHL1A----HVHOA AND SAHSR2-----HSROA (dependence between failure to swap to recirculation mode and failure to open dropline for DHR)  The basic event probability was changed from 2.00E-04 to 0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 15 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.26E-05 31.65 $109,449 Percent Change
-4.6% -2.9% -2.5% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 15 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.55E-07 1.56E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 6.78E-071.63E-072.07E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.60 8.92 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.17 1.00 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,649 $43,368 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $13,696 $3,293 $196
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-150 Three Mile Island Nuclear Station Unit 1 License Renewal Application Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-07 2.75E-09 7.45E-07 2.89E-07 3.19E-06 1.32E-05 1.69E-08 2.36E-06 1.91E-08 2.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 7.99E-08 1.43E-08 2.12E-07 2.75E-09 7.44E-07 2.89E-07 3.14E-06 1.23E-05 1.67E-08 2.33E-06 1.91E-08 2.26E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.29 0.00 1.00 0.39 6.97 3.28 0.00 0.62 0.01 31.65 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $755 $135 $810 $11 $2,842 $1,104 $19,719 $3,223 $4 $610 $5 $109,449 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 15 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,172,617 $99,094 2.0 $198,188 E.6.15.2 EXTERNAL FLOODING EVALUATION This SAMA is of importance in LOCA events when the entire volume of the BWST has been injected and the only source of borated water for continued core cooling is the water that has collected in the containment sump. For the external flooding cases, this is not an issue as the reactor is tripped by a manual shutdown (or a LOOP) rather than a LOCA event. While LOCAs are likely in external flooding scenarios due to SBO induced seal LOCAs, the primary side injection systems would be unavailable in those cases during the SBO. There is some potential for this SAMA to provide a benefit in the flood induced SBO scenarios where AC power is recovered prior to core damage, but the likelihood of recovering power in the short amount of time to prevent core damage is very low for flood conditions and this SAMA will provide an extremely limited benefit. To investigate this further, the SBO sequences from the base model were quantified and the resulting cutsets were reviewed to determine the contribution of manual recirculation alignment failures after power recovery. The sequences quantified included those in which AC power was both recovered and not recovered, specifically:
* LOOP-055
* LOOP-057 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-151 License Renewal Application
* LOOP-058
* LOOP-059
* LOOP-062
* LOOP-064
* LOOP-066
* LOOP-067
* LOOP-068
* LOOP-069 The only event identified in the cutsets was the JHEP event "JHHHL1AHSR2HEPOA" which accounted for only 0.1 percent of the SBO CDF. For the external flooding cases, the only two potential sequences that could be impacted by SAMA 15 are:
* Floods between 305' and 310' msl, sequence E:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (the SBO EDG is flooded) or because a flood warning is not provided and no preparations are made for the flood (the SBO EDG is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. These SBO cases are considered to be similar to the internal events SBO cases.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. While only a fraction of these cases would actually be SBOs, they are assumed to be 100% SBO cases for this evaluation. Assuming that SAMA 15 can remove all of the 0.1 percent risk attributed to manual recirculation failures results in a 0.1 percent reduction of the sequences identified above. The change in risk is trivial compared with the overall external flooding contributions, as summarized below: 
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-152 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 15 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.15 $542,125 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 15 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.65E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.70 0.25 0.37 177.15 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,826 $778 $1,217 $542,125 The external flooding component of the averted cost-risk for this SAMA is, therefore, $910: SAMA 15 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,542,563 $910 E.6.15.3 COST OF IMPLEMENTATION Multiple SAMA analyses have included estimates for this type of change, but the estimates vary by over a factor of 3.5:
* Oconee estimated the cost at over $1 million per unit (Duke 1998)
* Point Beach estimated the cost at over $1 million per unit (NMC 2004)
* Catawba estimated the cost at over $1 million (Duke 2001)
* Turkey Point estimated the cost to be about $450,000 (per unit) (FPL 2000)
* H.B. Robinson $265,000 (single unit) (CPL 2002)
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-153 License Renewal Application For TMI-1, the $450,000 estimate from Turkey Point is used as it is in the middle range of the industry estimates identified. E.6.15.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 15 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $198,188 $910 $199,098 $450,000 -$250,902 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.16 SAMA NUMBER 16:  AUT OMATE HPI INJECTION ON LOW PRESSURIZER LEVEL Providing an automatic signal to initiate HPI on low pressurizer level would improve the reliability of HPI initiation. The current initiation logic will not start HPI until low pressure (1600 psig) is reached in the RCS or high reactor building pressure (4 psig) is registered. This is adequate for LOCAs where the pressure drops with RCS level, but for loss of secondary side heat removal cases where the RCS pressure remains high while the level falls, no automated signal is available. HPI initiation is not a complicated action, but high workloads can divert attention from required tasks and providing an automated response to reduced level would prevent core uncovery in the event that a manual initiation is not performed. Pressurizer level instrumentation already exists for other purposes and the low level signal could be used as a means to start the HPI system. E.6.16.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events Environmental Report Appendix E  SAMA ANALYSIS Page E-154 Three Mile Island Nuclear Station Unit 1 License Renewal Application averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, cutset changes were made to address the impact of automating initiation of the HPI system. Thi s method was chosen given that HPI initiation reliability can easily be modified through the manipulation of existing human failure events. In the TMI-1 model, the relevant basic events include an independent event as well as joint human error events. In this case, automating the initiation of HPI (with human backup) is considered to reduce the failure probability to at least 1.0E-04, which is reflected by changing the failure probability of the independent HEP from 2.18E-03 to 1.0E-04. Because automation of the function basically removes it from the joint human error events, those events are set to 0.0. Any combinations of the remaining actions important to the model are treated in separate events and the development of new combinations is not required. The following table summarizes the model changes that were made: SAMA 16 - Model Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE BWHBW1-----HP2OA: OPERATOR FAILS TO INITIATE HPI The basic event probability was changed from 2.18E-03 to 1.00E-04. JHHMR1-HBW1HEPOA: MRHMR1-----HMUOA AND BWHBW1-----HP2OA (dependence between failure to initiate HPI and failure to establish a min flow path for the HPI pumps)  The basic event probability was changed from 1.40E-03 to 0.0. JHHAMHEFHBWHEPOA:  JHHAM2-HEF1HEPOA AND BWHBW1-----HP2OA(dependence between failure to initiate HPI, failure to start IA on emergency power, and failure to operate EF-V-30 locally after loss of IA) The basic event probability was changed from 2.40E-04 to 0.0. JHHEF1-HBW1HEPOA: EFHEF1_OPERH2HOA AND BWHBW1-----HP2OA (dependence between failure to initiate HPI and failure to locally operate the EFW flow control valves)The basic event probability was changed from 1.00E-04 to 0.0. JHHEF3-HBW1HEPOA: EFHEF3_OPERH2HOA AND BWHBW1-----HP2OA (dependence between failure to initiate HPI and failure to locally operate the EFW flow control valves after 2 hour bottle depletion) The basic event probability was changed from 4.10E-04 to 0.0.
JHHEF8-HBW1HEPOA: EFHEF8_OPERHBVOA AND BWHBW1-----HP2OA (dependence between failure to initiate HPI and failure to close EFW flow control block valve) The basic event probability was changed from 5.70E-05 to 0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-155 License Renewal Application SAMA 16 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.24E-05 24.27 $78,253 Percent Change
-5.5% -25.6% -30.3% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 16 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.55E-07 8.66E-07 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 6.93E-081.64E-072.13E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.60 4.95 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 0.43 1.01 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,649 $24,075 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $1,400 $3,313 $201 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.17E-072.69E-097.44E-072.89E-073.15E-061.34E-054.40E-09 2.34E-06 1.89E-082.24E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.29 0.00 1.00 0.39 6.99 3.58 0.00 0.62 0.01 24.27 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $829 $10 $2,842 $1,104 $19,782 $3,509 $1 $613 $5 $78,253 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 16 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,476,374 $795,337 2.0 $1,590,674
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-156 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.16.2 EXTERNAL FLOODING EVALUATION This SAMA is of importance primarily in loss of secondary side heat removal cases where low level can occur in the primary side without an RCS low pressure signal. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, all safety equipment is flooded and this SAMA has no impact on the risk.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (the SBO EDG is flooded) or because a flood warning is not provided and no preparations are made for the flood (the SBO EDG is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. In these cases, an SBO will occur that will lead to seal damage, the majority of which will be of the larger size leaks.
For these leaks, loss of inventory through the break will eventually result in a low pressure signal and an automatic HPI initiation if power is recovered. No benefit is considered available for these cases.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. For this evaluation, it is assumed that these sequences are impacted in the same manner as the internal events sequences, which are primarily loss of secondary side heat removal cases. The CDF for this flood sequence is reduced by the same percent as the internal events CDF based on SAMA 16 implementation. The following tables summarize the results of these changes on external flooding risk:  SAMA 16 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.10E-05 177.14 $542,092 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-157 License Renewal Application SAMA 16 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.36E-07 8.10E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.35 177.14 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,152 $542,092 The external flooding component of the averted cost-risk for this SAMA is summarized below: SAMA 16 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,541,516 $1,957 E.6.16.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $1,100,000 by the TMI staff (Exelon 2007c). E.6.16.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 16 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $1,590,674 $1,957 $1,592,631 $1,100,000 $492,631 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net
 
value is positive.
Environmental Report Appendix E  SAMA ANALYSIS Page E-158 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.17 SAMA NUMBER 17:  AUTO IS OLATE STEAM GENERATORS ON HIGH STEAM LINE FLOW For steam line breaks downstream of the MSIVs, failure to isolate the relevant steam generator is an important contributor to core damage. The addition of logic to isolate the steam generator on high steam line flow would reduce the core damage contribution from isolation failures. The steam line break contributors for TMI typically include multiple operator actions such that further procedure changes to direct mitigation of the event will have a limited impact due to operator dependence issues. The most effective solution was considered to be automation of a mitigating function. For the steam line break contributors, auto isolation of the MSIV was a straightforward change with the potential to impact a majority of the postulated scenarios. E.6.17.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate the automatic isolation of steam generators during a steamline break scenario, the HEP event SIHSI1-----HSGOA was reduced by a factor of 10 and any associated JHEP events set to 0.0 using previously generated cutsets. The new cutset probabilities for CDF and the various release categories were then summed and used to determine an estimate for the averted cost risk. Therefore, no new logic changes were made to the PRA model and no fault tree requantifications were performed. The following table summarizes the model changes that were made: SAMA 17 - Model Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE SIHSI1-----HSGOA: OPERATOR ERROR TO ISOLATE OTSG (BREAK DOWNSTREAM MSIV) The basic event probability was changed from 1.50E-02 to 1.50E-03. JHHSI1-HEF3HEPOA: SIHSI1-----HSGOA AND EFHEF3_OPERH2HOA (dependence between break isolation and failure to locally operate EF-V-30 after 2 hour air bottle depletion) The basic event probability was changed from 1.50E-02 to 0.0.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-159 License Renewal Application The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 17 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.34E-05 32.37 $111,518 Percent Change
-1.3% -0.7% -0.7% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 17 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.58E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.24E-071.65E-072.19E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.04 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.45 1.01 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $43,924 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,625 $3,333 $207 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.23E-072.75E-097.45E-072.89E-073.18E-061.30E-051.68E-08 2.35E-06 1.91E-082.34E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.06 3.46 0.00 0.63 0.01 32.37 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $852 $11 $2,846 $1,104 $19,970 $3,395 $4 $616 $5 $111,518 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 17 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,245,717 $25,994 2.0 $51,988
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-160 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.17.2 EXTERNAL FLOODING EVALUATION This SAMA does not have an impact on external flooding given that it impacts only steam line break initiating events. For the external flooding cases, this is not an issue as the reactor is tripped by a manual shutdown (or a LOOP) rather than a steam line break event. No measurable risk reduction is believed to result from implementation of this SAMA for external flooding, as shown below:  SAMA 17 - External Flooding Results CDF (/YR) DOSE-RISK (PERSON-REM/YR) OECR ($/YR) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 17 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E
-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E
-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 The external flooding component of the averted cost-risk for this SAMA is, therefore, $0: SAMA 17 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.17.3 COST OF IMPLEMENTATION This SAMA is considered to be similar in scope to SAMA 13 and the same cost of implementation ($950,000) is used for this SAMA.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-161 License Renewal Application E.6.17.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 17 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $51,988 $0 $51,988 $950,000 -$898,012 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.18 SAMA NUMBER 18:  PROVID E THE CAPABILITY TO ALIGN THE STANDBY BATTERY CHA RGER AND THE 1A/1B DC CROSS-TIE FROM THE MCR TMI has a spare 125V DC battery charger for each division that can be aligned to either battery bank within a division in the event that a normally operating battery charger fails. Currently, the alignment requires local actions. There is typically adequate time to align the charger in the event of a failure given that the batteries will last at least four hours, but additional changes could be made to allow rapid alignment of the spare charger from the MCR to reduce the manipulation time and improve the man-machine interface. A divisional cross-tie exists that can be used to tie the DC buses together, if required. Providing the capability to remotely operate the cross-tie would provide an additional means of maintaining DC power to required loads. E.6.18.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA.
Environmental Report Appendix E  SAMA ANALYSIS Page E-162 Three Mile Island Nuclear Station Unit 1 License Renewal Application To simulate alignment of a spare battery charger from the MCR, the HEP event DABATTCHGR-HBCOA, which assumed manipulations performed outside the MCR, was lowered by a factor of 10. There were no applicable JHEP events; therefore, no additional basic event data changes were required. No changes were made to the cutsets to explicitly represent the improvements to the cross-division DC cross-tie, but not modeling this capability does not have a meaningful impact on the results for the following reasons:
* The baseline model does not credit the existing, proceduralized action to cross-tie the DC buses. Given that there is ample time to perform the cross-tie, the base model over-emphasizes the importance of the DC power supplies.
* The HEP representing alignment of the spare battery chargers (DABATTCHGR-HBCOA) is currently assigned a screening value of 0.1. Like the DC cross-tie, spare battery charger alignment is proceduralized and ample time is available for completing the action. If reasonable credit was assigned to DABATTCHGR-HBCOA, the importance of the DC power supplies would be reduced.
* Even with the low credit for DABATTCHGR-HB COA, the RRW for the action is only 1.001 when SAMA 18 is implemented. This implies that further reductions to the DC power supplied would provide limited benefit. As a result, the changes made to HEP DABATTCHGR-HBCOA are considered to provide a reasonable assessment of the benefits related to SAMA 18. The following table summarizes the changes that were made to the cutsets: SAMA 18 - Model Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE DABATTCHGR-HBCOA:  The basic event probability was changed from 1.00E-01 to 1.00E-02. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-163 License Renewal Application SAMA 18 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.33E-05 32.54 $112,239 Percent Change
-1.7% -0.2% -0.0% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 18 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.24E-08 9.07E-119.07E-111.90E-102.88E-103.64E-081.35E-088.54E-09 2.87E-07 7.25E-071.65E-072.19E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.84 4.46 1.01 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $231 $3 $3 $7 $11 $327 $121 $77 $2,580 $14,645 $3,333 $207 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 2.79E-08 2.88E-075.41E-097.86E-073.71E-073.17E-061.25E-051.69E-08 2.55E-06 1.91E-082.33E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.08 0.39 0.01 1.06 0.50 7.04 3.33 0.00 0.68 0.01 32.54 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $264 $1,100 $21 $3,003 $1,417 $19,908 $3,272 $4 $668 $5 $112,239 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 18 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,259,138 $12,573 2.0 $25,146
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-164 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.18.2 EXTERNAL FLOODING EVALUATION This SAMA is potentially of importance in any event where power to the battery chargers is available. For the external flooding cases, the only two potential sequences that could be impacted by SAMA 18 are:
* Floods between 305' and 310' msl, sequence E:  Most of the sequences in the 305' to 310' msl range are not impacted by this SAMA as core damage is caused by failure of the flood gates (all safety equipment is flooded) or because a flood warning is not provided and no preparations are made for the flood (all safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of onsite power leads to core damage. This SAMA would provide benefit for flood sequence "E" when 1) battery depletion is the eventual cause of onsite power failure and alignment of the standby charger would prevent loss of DC power, and 2) when offsite AC power is recovered after loss of all on-site AC power and alignment of the standby charger would restore DC power. 
 
For this evaluation, the characteristics of the SBO contributors in flooding sequence "E" are assumed to be the same as the internal events SBO contributors. This is considered to be reasonable given that the flood gates prevent damage to plant safety equipment and offsite power recovery is a minor contributor to the internal events SBO evolutions (implies the potentially longer offsite AC recovery times for flood events are not a factor).
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. For LOOP cases, improved DC reliability can
 
impact the CDF. As mentioned above, there is some potential for this SAMA to provide a benefit in the flood induced SBO scenarios (flood sequence "E"), but the circumstances in which the SAMA could be used are rare and it will provide an extremely limited benefit. To investigate this further, the SBO sequences from the base model were quantified and the resulting cutsets were reviewed to determine the contribution of manual alignment of the spare battery chargers. The sequences quantified included those in which AC power was both recovered and not recovered, specifically:
* LOOP-055
* LOOP-057 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-165 License Renewal Application
* LOOP-058
* LOOP-059
* LOOP-062
* LOOP-064
* LOOP-066
* LOOP-067
* LOOP-068
* LOOP-069 Basic event "DABATTCHGR-HBCOA" accounted for only 0.8 percent of the SBO CDF, which implies that of all SBO cases, only 0.8 percent of the contribution includes conditions in which SAMA 18 could provide any benefit. Assuming that SAMA 18 can remove all of the 0.8 percent of the risk attributed to manual battery charger alignment failures results in a 0.8 percent reduction of 305' to 310' flood sequence E CDF. For external floods below 305' mls, the impact could be larger than for SBO scenarios given that that the need to recover or retain some form of AC power is not a precondition for credit (AC power is already available to the chargers). In these cases, the CDF is considered to behave more like the overall internal events model rather than the SBO subset of the CDF. To represent this behavior, the CDF for external floods below 305' msl is reduced in proportion to the internal events model based on SAMA 18 implementation. The results of these processes are summarized below:  SAMA 18 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.10E-05 177.06 $541,876 Percent Change 0.0% 0.1% 0.1%
Environmental Report Appendix E  SAMA ANALYSIS Page E-166 Three Mile Island Nuclear Station Unit 1 License Renewal Application A further breakdown of this information is provided below according to release category. SAMA 18 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.63E-06 8.65E-08 2.46E-07 8.10E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.62 0.25 0.36 177.06 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,596 $778 $1,1198 $541,876 The external flooding component of the averted cost-risk for this SAMA is summarized below: SAMA 18 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,535,359 $8,114 E.6.18.3 COST OF IMPLEMENTATION No plant specific implementation cost was developed for this SAMA. Based on the low impact of the SAMA, the $100,000 minimum cost of a hardware modification (Exelon 2003) is used as the implementation cost. E.6.18.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 18 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $25,146 $8,114 $33,260 $100,000 -$66,740 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-167 License Renewal Application E.6.19 SAMA NUMBER 19:  INSTALL BATTERY BACKED HYDROGEN IGNITORS OR A PASSIVE HY DROGEN IGNITION SYSTEM The addition of hydrogen igniters would provide a means of preventing catastrophic combustible gas burns, which may lead to containment failure, by continuously burning these gases before they reach critical levels. Providing battery backup power would increase the likelihood that this system would be available in LOOP events. Use of a passive system would also function in LOOP as well as long term SBO scenarios. E.6.19.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate installation of hydrogen ignitors, a new basic event (SAMA19-H2IGNITER) was created to simulate the installation of a proposed hydrogen ignition system to minimize the concentration of hydrogen buildup within containment from various hydrogen producing mechanisms, such as corium-concrete interaction. Addition of this basic event to the Level 2 model necessitated inserting a new level of logic above the existing gate H2BURNS. Specifically, a new AND gate named H2BURNS-1 was inserted as an input to the Containment Event Tree nodal top event EARLY. The two inputs to gate H2BURNS-1 are gate H2BURNS and the new basic event SAMA19-H2IGNITER, with an assumed unavailability of 1.0E-02. This estimate is based on estimating an overall unavailability of a proposed system without identifying any particular design features or support dependencies (consistent with a passive design or an independent battery support system), and also represents a number that is not overly conservative or one that would tend to exaggerate the success of such a proposed system. In addition, hydrogen burns are potential contributors to containment late; however, review of the cutsets shows that these evolutions are probabilistically insignificant (no cutsets exist that include late containment failure cause by hydrogen burns). All accident sequences including late hydrogen burns result in an intact containment and hydrogen igniters would not impact the Environmental Report Appendix E  SAMA ANALYSIS Page E-168 Three Mile Island Nuclear Station Unit 1 License Renewal Application results. Consequently, no model changes were included in this quantification to address late hydrogen burns to simplify the modeling process. The model changes identified above yielded a reduction in the Dose-risk and Offsite Economic cost-risk, but did not impact the CDF, as summarized below:  SAMA 19 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.37E-05 29.11 $100,376 Percent Change 0.0% -10.7% -10.6% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 19 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 1.76E-071.33E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 1.08 0.82 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $3,555 $2,687 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-07 2.75E-09 7.45E-07 2.89E-07 3.19E-06 1.32E-05 1.69E-08 2.36E-06 1.91E-08 2.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-07 2.75E-09 7.45E-07 2.89E-07 3.19E-06 1.38E-05 1.69E-08 2.36E-06 1.91E-08 2.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.69 0.00 0.63 0.01 29.11 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,617 $4 $618 $5 $100,376 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-169 License Renewal Application SAMA 19 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,987,716 $283,995 2.0 $567,990 E.6.19.2 EXTERNAL FLOODING EVALUATION This SAMA can impact many of the external flooding evolutions given that a passive hydrogen ignition system could be available even in extreme flooding conditions. The circumstances related to each flood range are discussed below:
* Floods over 310' msl:  For these floods, water level increases until it pours over the top of the existing flood barriers. Once core damage has occurred, the containment response is similar to an SBO scenario where water is not on the containment floor. The early containment failure frequency is assumed to be reduced in proportion to the early containment failures of the internal events model for this SAMA.
* Floods between 305' and 310' msl:  Scenarios "A" through "D" are the result of flood gate failures. In these scenarios, no credit would be available for those cases where containment isolation was successfully performed (containment isolation failure will remain as containment isolation failures). In scenarios "A" and "C", cold shutdown is achieved and containment isolation is successful, therefore credit is taken for these cases. For sequences "B" and "D", no credit can be taken for hydrogen ignitors as these sequences represent cases where transition to cold shutdown has not occurred and the containment is not isolated. In sequence "E", the flood gates hold, but EDG failures cause an SBO and prevent a transition to cold shutdown, which results in containment isolation failure and no credit is taken for SAMA 19. For sequence "F", the operators have no warning of the impending flood and the plant is also not transitioned to cold shutdown before flood damage occurs, which implies containment isolation failure and no credit for SAMA 19.
* Floods below 305' mls:  These are similar to internal events LOOP scenarios and early containment failures are assumed to be reduced in proportion to those in the internal events
 
model. Based on the qualitative descriptions above, the following quantitative structure was developed to represent the implementation of this SAMA:
Environmental Report Appendix E  SAMA ANALYSIS Page E-170 Three Mile Island Nuclear Station Unit 1 License Renewal Application External Flood Sequence Identifier Quantification Method >310 Feet Reduce the "EARLY" release category (RC5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Late-SM" release category (RC7) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 7. 305 to 310 feet Sequence "A" Reduce the "EARLY" release category (RC5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Late-SM" release category (RC7) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 7. 305 to 310 feet Sequence "B" No change is made to this sequence's distribution. 305 to 310 feet Sequence "C" Reduce the "EARLY" release category (RC5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Late-SM" release category (RC7) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 7. 305 to 310 feet Sequence "D" No change is made to this sequence's distribution. 305 to 310 feet Sequence "E" No change is made to this sequence's distribution. 305 to 310 feet Sequence "F" No change is made to this sequence's distribution. <305 feet Reduce the "EARLY" release category (RC5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Late-SM" release category (RC7) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 7. Due to the relatively large early containment failure component of the external events model, this SAMA has a large impact on the external flooding risk, as summarized below:  SAMA 19 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 145.71 $434,849 Percent Change 0.0% -17.8% -19.8% A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-171 License Renewal Application SAMA 19 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 104.56 10.34 0.19 1.49 17.87 10.71 0.25 0.30 145.71 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $309,750 $30,635 $598 $4,401 $54,839 $32,858 $778 $991 $434,849 The corresponding external flooding component of the averted cost-risk is shown below: SAMA 19 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $12,983,587 $2,559,886 E.6.19.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $760,000 in the Calvert Cliffs SAMA analysis (BGE 1998). E.6.19.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 19 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $567,990 $2,559,886 $3,127,876 $760,000 $2,367,876 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive.
Environmental Report Appendix E  SAMA ANALYSIS Page E-172 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.20 SAMA NUMBER 20:  EXTEND THE HIGH PRESSURE BOUNDARY THROUGH DHR VALVE DH-V-3 FOR ISLOCA ISOLATION The highest frequency ISLOCA core damage scenario for TMI-1 is through two valves in the DHR suction line. While the frequency is relatively low in terms of CDF, the release frequency is relatively high given that primary containment is bypassed by definition. No effective mitigating actions are considered to be available in these cases because 1) the break may occur upstream of DH-V-3 or additional breaks in the low pressure boundary may occur after closure of a low pressure isolation valve, 2) reduction of primary system pressure may reduce the flow out of the break, but it would not stop it, and 3) refill of the BWST does not place the plant in a stable state and the impacts of auxiliary building flooding would have to be addressed before a successful endstate could be assigned to this type of action. Extending the pressure boundary through DH-V-3 would provide an additional isolation point in these cases. This SAMA would provide an effective means of terminating the ISLOCA event and the reliability would be limited primarily by the ability of the operators to diagnose the event. Maintaining DH-V-3 as a motor operated valve will ensure that the break can be isolated quickly and without exposing the operators to potentially hazardous conditions. E.6.20.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of SAMA 20's averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, cutset changes were made to address the impact of extending the high pressure boundary of the DHR suction line. This method was chosen given
 
that ISLOCA events are modeled in a single cutset and are easily manipulated within the cutsets. While the lumped event includes more than one ISLOCA contributor, most of the risk is due to the DHR suction line scenario, so it is assumed that manipulation of the ISLOCA cutset can be used to represent changes to the DHR suction line scenario frequency. For the purposes of this analysis, implementation of this SAMA is assumed to eliminate ISLOCA risk completely. The following table summarizes the changes that were made to the cutsets:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-173 License Renewal Application SAMA 20 - Cutset Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE %ISL:  INTERFACING SYSTEM LOCA The initiating event probability was changed from 1.80E-07 to 0.0. The change identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 20 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.35E-05 31.65 $108,733 Percent Change
-0.8% -2.9% -3.1% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 20 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 6.31E-10 3.66E-09 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.00 0.02 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $12 $68 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.35E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.52 0.00 0.63 0.01 31.65 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,459 $4 $618 $5 $108,733 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-174 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 20 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,184,724 $86,987 2.0 $173,974 E.6.20.2 EXTERNAL FLOODING EVALUATION This SAMA does not have an impact on external flooding given that it impacts only ISLOCA events and dual of importance in SGTR events where RCS inventory leaves the containment and is unavailable for recirculation from the sump. For the external flooding cases, this is not an issue as the reactor is tripped by a manual shutdown (or LOOP) rather than an ISLOCA event.
While LOCAs are likely in external flooding scenarios due to SBO induced seal LOCAs, the sump would be available if AC power was subsequently recovered. No measurable risk reduction is believed to result from implementation of this SAMA for external flooding, as summarized below:  SAMA 20 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,159 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. SAMA 20 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 1.33E+02 1.31E+01 1.90E-01 1.89E+00 1.79E+01 1.07E+01 2.50E-01 3.70E-01 1.77E+02 SAMA Dose-Risk 1.33E+02 1.31E+01 1.90E-01 1.89E+00 1.79E+01 1.07E+01 2.50E-01 3.70E-01 1.77E+02 Base OECR 4.06E+05 4.01E+04 5.98E+02 5.77E+03 5.48E+04 3.29E+04 7.78E+02 1.22E+03 $542,159 SAMA OECR 4.06E+05 4.01E+04 5.98E+02 5.77E+03 5.48E+04 3.29E+04 7.78E+02 1.22E+03 $542,159 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-175 License Renewal Application The external flooding component of the averted cost-risk for this SAMA is, therefore, $0: SAMA 20 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,473 $0 E.6.20.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $3,030,000 by the TMI staff (Exelon 2007c). E.6.20.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 20 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $173,974 $0 $173,974 $3,030,000 -$2,856,026 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.21 SAMA NUMBER 21:  INSTALL CONCRETE SHIELDS TO BLOCK DIRECT PATHWAYS FROM THE RPV TO THE CONTAINMENT WALL AND/OR DIRECT CONTAINMENT FLOODING EARLY IN EXTERNAL FLOODING SCENARIOS This SAMA is based on a failure mode identified in the Level 2 analysis that indicates core debris ejection during reactor vessel failure could result in dispersal of debris such that it could directly interact with the containment wall and cause a failure of the wall (early containment failure). Quantitatively, the largest contributor comes from low pressure melt cases where the core debris flows over the containment floor to contact the containment wall. This type of interaction could be prevented through the installation of concrete barriers to contain the core
 
debris away from the outer containment wall.
Environmental Report Appendix E  SAMA ANALYSIS Page E-176 Three Mile Island Nuclear Station Unit 1 License Renewal Application Another option for this SAMA, which is important for external flooding cases, is to direct flooding of the containment early so that water would be on the floor of the containment before core damage/vessel failure. For internal events evolutions, the SAMGs direct containment flooding when there are indicators of the onset of core damage (e.g., high core temperatures, hydrogen in the reactor building), which adequately addresses the sequences of concern. For external flooding cases, however, the ability to initiate containment sprays will be lost before there are any indicators of core damage such that the existing SAMGs cannot be credited for directing containment flooding. Both the installation of concrete barriers and the procedure changes for external flooding cases are discussed in more detail in the following subsections. E.6.21.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component this SAMA's averted cost-risk associated with the internal events and the non-external flooding external events. As described
 
in Section E.4.6.3 , the e x ternal eve n ts risk, excl u ding external flooding, is considered t o be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. In order to represent this SAMA in the PRA, cutset changes were made to address the impact of installing the concrete barriers. This method was chosen given that the containment wall/core debris interaction events are represented by two basic events and can easily be changed. In the TMI-1 model, there is a low pressure melt case (CWNOLIMITLPME) and a high pressure melt case (CWNOLIMITHPME). The low pressure melt case is by far the more significant contributor of the two to the early containment failure frequency. The high pressure melt case, while already a low contributor, is linked to the failure to locally operate the ADVs, which is currently assigned a value of 1.0. Procedures exist at TMI-1 to operate the ADVs locally, but the model does not currently credit the procedures. As a result, the importance of CWNOLIMITHPME is artificially inflated. CWNOLIMITHPME could be excluded from consideration in this SAMA, but for completeness, both CWNOLIMITHPME and CWNOLIMITLPME are included in the modeling changes. While this SAMA does reduce the early containment failure frequency, it does not necessarily eliminate the release and it must be re-distributed to prevent over crediting this SAMA. The Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-177 License Renewal Application concrete barrier will prevent core debris attack on the containment wall, but basemat failure could still occur depending on the availability of water on the containment floor and the coolability of the core debris. Based on a review of the cutsets containing events CWNOLIMITLPME and CWNOLIMITHPME, containment spray is available about 50 percent of the time. For the cases where it is not available, basemat failure is assumed. When containment spray is available, the debris is assumed to be coolable only 50 percent of the time, which is consistent with the TMI-1 Level 2 analysis. When spray is containment spray is available and the debris is coolable, it is assu med that containment heat removal is available and that these cases result in an intact containment (RC group 9) rather than a late overpressurization failure (RC group 7). The following table summarizes the model changes that were made: SAMA 21 - Model Changes GATE AND / OR BASIC EVENT ID  AND DESCRIPTION DESCRIPTION OF CHANGE CWNOLIMITLPME: Plant Config and Layout Does Not Limit Material Reaching Cont. Wall With LPM The basic event probability wa s changed from 1.00E-01 to 0.0 to account for the ability of the concrete barriers to prevent failure of the containment wall. CWNOLIMITHPME: Plant Config and Layout Does Not Limit Material Reaching Cont. Wall With HPM The basic event probability wa s changed from 1.00E-01 to 0.0 to account for the ability of the concrete barriers to prevent failure of the containment wall. RC8 (Basemat failure) Frequency  Increase the frequency by 75percent of the reduction in RC5 (Early containment failure). This accounts for both the cases in which containment spray is not available and those cases where containment spray is available, but the debris is not coolable.  (0.5
* RC5 reduction + 0.5
* RC5 reduction
* 0.5 =
0.75
* RC5 reduction) RC9 (Intact) Frequency Increase the frequency by 25 percent of the reduction in RC5 (Early containment failure). This accounts for the cases in which containment spray is available and the core debris is coolable. (0.5
* RC5 reduction
* 0.5 = 0.25
* RC5 reduction) The model changes identified above yielded no change in the CDF, but did reduce the Dose-risk and Offsite Economic cost-risk, as summarized below:  SAMA 21 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.37E-05 31.5 $108,333 Percent Change 0.0% -3.4% -3.5%
Environmental Report Appendix E  SAMA ANALYSIS Page E-178 Three Mile Island Nuclear Station Unit 1 License Renewal Application A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 21 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 5.94E-075.65E-082.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 3.65 0.35 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $11,999 $1,141 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.38E-061.32E-058.05E-08 2.36E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.51 3.53 0.02 0.63 0.01 31.50 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $21,232 $3,461 $21 $618 $5 $108,333 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 21 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,179,284 $92,427 2.0 $184,854 E.6.21.2 EXTERNAL FLOODING EVALUATION The typical external flooding evolution is one in which the plant is stable until flood waters breach the flood gates and fail safety equipment. Alone, the concrete barriers would shift containment failure and the corresponding release from the "Early" release category (RC group
: 5) to the "basemat failure" category (RC group 8), assuming the containment is isolated. Flooding the containment floor can prevent core/concrete attack 50 percent of the time and in conjunction with the concrete barriers, the containment failure mode would be a long term Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-179 License Renewal Application overpressurization failure with a release that would be categorizes as "Late-small" (RC group 8). For the internal events model, an "intact" cont ainment state was assumed to be possible given the potential for heat removal to be available, but for the external flooding cases, heat removal is not assumed to be available due to SBO conditions and a late overpressurization failure is considered to be more appropriate. For the 50 percent of the cases in which core-concrete attack is not prevented, basemat failure is assumed (RC group 8). For many flood sequences, the loss of AC power will not be anticipated in time to initiate containment flooding, but in some cases, changes in the procedures could allow containment flooding as a means of reducing the release severity:
* Floods over 310' msl:  For these floods, water level increases until it pours over the top of the existing flood barriers. In these evolutions, there would likely be an interval when water level is rising between 305' and 310' msl when the determination could be made that the flood water will eventually top the barriers and that containment flooding should be performed as a precaution. Given that the flood gates are available and can be used to maintain the core in a safe state without risking further damage to the plant, flooding the containment floor to a depth where the water would remain available until vessel breach would be undesirable until absolutely necessary. While this is true, it is a credible means of reducing the probability of early containment failure and basemat failures. Containment spray is assumed to always be available before flood gate topping such that containment flooding will be successful, if directed.
* Floods between 305' and 310' msl:  Scenarios "A" through "D" are the result of flood gate failures. In these scenarios, no credit would be available for performing early containment spray as core damage would not be anticipated and there would be no desirable cue to direct containment flooding. In scenarios "A" and "C", credit could be taken for the concrete barriers as cold shutdown is achieved (implies containment isolation) and the barriers would prevent interaction with the containment wall. For sequences "B" and "D", no credit can be taken for the concrete barriers as these sequences represent cases where transition to cold shutdown has not occurred and the containment is not isolated (these remain containment isolation failure cases). In sequence "E", the flood gates hold, but EDG failures cause an SBO and no credit would be available for containment spray. Due to the AC power failures, the plant was not transitioned to cold shutdown and a containment isolation failure would occur (no credit for concrete barriers). For sequence "F", the operators have no warning of Environmental Report Appendix E  SAMA ANALYSIS Page E-180 Three Mile Island Nuclear Station Unit 1 License Renewal Application the impending flood and the plant is also not transitioned to cold shutdown before flood damage occurs, which implies containment isolation failure.
* Floods below 305' mls:  These are similar to LOOP scenarios and containment flooding is already addressed by the SAMGs. These cases are treated in the same manner as the internal events cases. Based on the qualitative descriptions above, the following quantitative structure was developed to represent the implementation of this SAMA: External Flood Sequence Identifier Quantification Method >310 Feet
* Reduce the "EARLY" release category (RC group 5) frequency contribution by the same fraction that this SAMA reduced the internal events RC group 5 frequency.
* Increase the basemat failure frequency (RC group 8) by 50 percent of the reduction in RC group 5. This accounts for the cases where containment spray is available, but the debris is not coolable. Containment spray is assumed to always be available before topping of the flood gates.  (0.5
* RC5 reduction)
* Increase the late containment failure frequency (RC group 7) by 50 percent of the reduction in RC group 5. This accounts for the cases in which containment spray is available, the core debris is coolable, and lack of heat removal results in late containment failure. Containment spray is assumed to always be available before topping of the flood gates.  (0.5
* RC5 reduction) 305 to 310 feet Sequence "A" Reduce the "EARLY" release category (RC group 5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Basemat Failure" release category (RC8) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 8. 305 to 310 feet Sequence "B" No change is made to this sequence's distribution. 305 to 310 feet Sequence "C" Reduce the "EARLY" release category (RC5) frequency contribution by the same fraction that this SAMA reduced the internal events RC5 frequency. Increase the "Basemat Failure" release category (RC8) frequency by the amount this SAMA reduced the RC5 frequency to simulate the shift of the release from RC group 5 to RC group 8. 305 to 310 feet Sequence "D" No change is made to this sequence's distribution. 305 to 310 feet Sequence "E" No change is made to this sequence's distribution. 305 to 310 feet Sequence "F" No change is made to this sequence's distribution.
<305 feet Treated in the same manner as the internal events model.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-181 License Renewal Application Due to the relatively large early containment failure component of the external events model, this SAMA has a large impact on the external flooding risk, as summarized below:  SAMA 21 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 165.06 $500,115 Percent Change 0.0% -6.8% -7.8% A further breakdown of this information is provided below according to release category. SAMA 21 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 121.80 12.15 0.19 1.75 17.87 10.71 0.25 0.34 165.06 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $367,957 $36,696 $598 $5,271 $54,839 $32,858 $778 $1,118 $500,115 The external flooding component of the averted cost-risk is summarized below: SAMA 21 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $14,547,190 $1,181,137 E.6.21.3 COST OF IMPLEMENTATION The cost of implementation is estimated to be $1,200,000 by the TMI staff (Exelon 2007c). E.6.21.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-182 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 21 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $184,854 $996,283 $1,181,137 $1,200,000 -$18,863 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.22 SAMA NUMBER 22:  INSTALL AN INDEPENDENT EFW SYSTEM For TMI-1, loss of MFW after a trip coupled with loss of EFW can lead to large radionuclide releases in SGTR and induced SGTR scenarios due to the unavailability of water in the SGs for fission product scrubbing. A large contributor to EFW failure is estimated to be system wide common cause failures. An independent, motor dr iven, auxiliary feedwater system would be an effective means of addressing these cases. Power dependence is not a large issue for the cases addressed by this SAMA and the independent EFW pump is assumed to be powered by existing emergency power such that it would not be capable of mitigating SBO scenarios. E.6.22.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. New simplified model logic was added to the PRA model to represent an independent system that provides a backup to the existing EFW system. Requantification of the PRA model was then performed to determine new core damage and release category frequencies. Specifically, for non-ATWS scenarios, a new level of logic was added above the gate EFNOATWS (EFW without ATWS conditions) consisting of a new AND gate named EFNOATWS-1 with two inputs. The two inputs to this new gate consisted of the original logic gate EFNOATWS and a new OR gate named ALT-EFW-NOATWS (failure of alternate EFW for non-ATWS conditions). The inputs to the OR gate ALT-EFW-NOATWS consisted of the following inputs:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-183 License Renewal Application SAMA 22 NON-ATWS BASIC EVENTS UNAVAILABILITY EVEN T DESCRIPTION SAMA22ELECNOATWS 1.00E-02 NON-ATWS ALTERNATE EFW ELECTRICAL POWER DEPENDENCY FAILURES SAMA22MECHNOATWS 1.00E-03 NON-ATWS ALTERNATE EFW MECHANICAL DEPENDENCY FAILURES SAMA22HEP-NOATWS 1.00E-01 NON-ATWS ALTERNATE EFW HEP FAILURES SAMA22JHEPNOATWS 5.00E-02 NON-ATWS ALTERNATE EFW JHEP FAILURES The electrical event unavailability was based on the assumption that electrical dependencies require several other dependencies and control signals to function properly, thus resulting in a higher unavailability relative to assumed mechanical failures. The mechanical unavailability event was arbitrarily represented as 0.001, since most mechanical failures are typically of this order of magnitude. The independent HEP event was arbitrarily assigned an unavailability of 0.1, since this was based on the assumption that several actions would need to be performed outside the MCR. The JHEP event was estimated as having a high dependence for failure of a second related event, meaning that failure of the second HEP event was highly dependent on failure of the HEP event SAMA22HEP-NOATWS. For ATWS scenarios, the fault tree logic for gate EFATWS was modified in the same manner as described above. In addition, the unavailabilities for the added basic events discussed above were increased by a factor of 3 (half a decade based on a logarithmic scale) to account for ATWS environmental stress factors and a greater sense of urgency. These events are identified in the table below:
SAMA 22 ATWS BASIC EVENTS UNAVAILABILITY EVEN T DESCRIPTION SAMA22ELEC--ATWS 3.00E-02 ATWS ALTERNATE EFW ELECTRICAL POWER DEPENDENCY FAILURES SAMA22MECH--ATWS 3.00E-03 ATWS ALTERNATE EFW MECHANICAL POWER DEPENDENCY FAILURES SAMA22HEP---ATWS 3.00E-01 ATWS ALTERNATE EFW HEP FAILURES SAMA22JHEP--ATWS 1.50E-01 ATWS ALTERNATE EFW JHEP FAILURES All affected logic described above that is modeled within the logic structure for post-LOOP recovery scenarios was also modified, with applicable gate names appended with the characters "-R". This was only necessary for the logic involving non-ATWS scenarios.
Environmental Report Appendix E  SAMA ANALYSIS Page E-184 Three Mile Island Nuclear Station Unit 1 License Renewal Application The model changes identified above yielded a reduction in the Dose-risk and Offsite Economic cost-risk, but did not impact the CDF, as summarized below:  SAMA 22 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.22E-05 27.05 $85,423 Percent Change
-6.3% -17.0% -23.9% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 22 Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.21E-07 6.49E-07 1.80E-07 9.45E-09 9.07E-119.07E-111.90E-103.46E-103.81E-082.43E-088.54E-09 5.12E-07 6.88E-071.70E-072.14E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.41 3.71 0.91 0.05 0.00 0.00 0.00 0.00 0.11 0.07 0.03 1.50 4.23 1.05 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $11,704 $18,042 $3,348 $176 $3 $3 $7 $13 $343 $218 $77 $4,603 $13,898 $3,434 $202 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.72E-10 0.00E+00 8.02E-08 1.49E-08 2.18E-072.86E-097.47E-072.90E-073.14E-061.26E-052.70E-09 2.34E-06 3.08E-092.22E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.23 0.04 0.29 0.00 1.01 0.39 6.97 3.37 0.00 0.62 0.00 27.05 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $3 $0 $758 $141 $833 $11 $2,854 $1,108 $19,719 $3,311 $1 $613 $1 $85,423 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 22 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $2,662,735 $608,976 2.0 $1,217,952
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-185 License Renewal Application E.6.22.2 EXTERNAL FLOODING EVALUATION This SAMA will have a limited impact for external flooding scenarios given that many of the scenarios result in flooding the plant's safe ty equipment, which would render the proposed equipment inoperable. Even if the independent EFW system were located in a flood safe zone, the floods cause an SBO and subsequent seal LOCA that would result in core damage regardless of the operability of an alternate EFW system. The circumstances related to each flood range are discussed below:
* Floods over 310' msl:  For these floods, water level increases until it pours over the top of the existing flood barriers. Flooding of safety equipment occurs and the subsequent seal LOCA will lead to core damage with or without SAMA 22. No credit is taken for this SAMA for this flood scenario.
* Floods between 305' and 310' msl:  Scenarios "A" through "D" are the result of flood gate failures. In these scenarios, the result is the same as for floods over 310' msl and no credit is taken for SAMA 22. In sequence "E", the flood gates hold, but EDG failures cause an SBO. Alternate EFW could be beneficial if AC power was recovered to provide primary side makeup, but review of the LOOP/SBO model and the baseline SBO cutsets shows that EFW operability is only important to prolonging the time to core damage to allow AC power recovery. For floods of this magnitude, the normal AC power recovery credits are not considered to be applicable and the benefit of delaying core damage for a matter of a couple of hours would be minimal. No credit is taken for this SAMA for sequence "E". For sequence "F", the operators have no warning of the impending flood and the flood gates are not installed in time to prevent flooding of the safety equipment. As with the other, similar sequences, no credit is taken for SAMA 22 for sequence "F".
* Floods below 305' mls:  These are similar to internal events LOOP scenarios and the availability of an alternate EFW system would be beneficial. In order to simplify the modeling for this sequence, SAMA 22 is assumed to eliminate all risk from these flood
 
scenarios. Implementation of this SAMA would result in only a limited risk reduction, as summarized below:
Environmental Report Appendix E  SAMA ANALYSIS Page E-186 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 22 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.08E-05 176.79 $540,940 Percent Change
-0.3% -0.2% -0.2% A further breakdown of this information is provided below according to release category. SAMA 22 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 0.00E+00 8.08E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.00 176.79 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $0 $540,940 The corresponding external flooding component of the averted cost-risk is shown below: SAMA 22 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,507,657 $35,816 E.6.22.3 COST OF IMPLEMENTATION Calvert Cliffs estimated the cost of installing an additional HPSI pump with a dedicated diesel to be between $5 million and $10 million. This type of enhancement is similar is scope to the changes required for this SAMA and the lower bound estimate of $5 million is used for this SAMA as the independent diesel is not required for this SAMA. E.6.22.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-187 License Renewal Application SAMA 22 - Net Value Non-External Flooding Based Averted Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $1,217,952 $35,816 $1,253,768 $5,000,000 -$3,746,232 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.23 SAMA NUMBER 23:  DEVELOP ALARM RESPONSE PROCEDURES TO DIRECT OPERATION OF RR-V-5 ON LOW RBEC FLOW Failure of RR-V-6 to open results in the loss of RBEC flow to the reactor building coolers, which can be diagnosed using the system flow indicators in the main control room; however, no alarm response procedures exist to specifically direct operation of the bypass valve (RR-V-5). If this procedure was developed, it may reduce the diagnosis time and improve the reliability of this operator action in an accident conditions. E.6.23.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMAs averted cost-risk associated with the internal events and the non-external flooding events. As described in Section E.4.6.3 , the external events risk, excl udi n g external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events based averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To model a more improved procedure regarding loss of RBEC flow from failure of RR-V-6 and recovery via MOV RR-V-5, the HEP event CFHRR1-----HVAOA was reduced by a factor of 10, from 7.79E-01 to 7.79E-02. There were no applicable JHEP events. The new cutset probabilities for the various Level 2 release categories were then summed and used to determine an estimate for the averted cost risk. Therefore, no new logic changes were made to the PRA model and no fault tree requantifications were performed. The following table summarizes the changes that were made to the model: SAMA 23 - Model Changes GATE AND / OR BASIC E VENT ID AND DESCRIPTION DESCRIPTION OF CHANGE CFHRR1-----HVAOA:
OPERATOR FAILS TO OPEN MOV RR-V-5 The basic event probability was changed from 7.79E-1 to 7.79E-02.
Environmental Report Appendix E  SAMA ANALYSIS Page E-188 Three Mile Island Nuclear Station Unit 1 License Renewal Application The model changes identified above yielded a reduction in the Dose-risk and Offsite Economic cost-risk, but did not impact the CDF, as summarized below:  SAMA 23 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.37E-05 32.42 $111,626 Percent Change 0.0% -0.6% -0.6% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 23 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-071.36E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.04 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $129 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 7.72E-08 1.43E-08 1.42E-072.69E-097.13E-072.89E-073.17E-061.34E-051.69E-08 2.34E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.19 0.00 0.96 0.39 7.04 3.57 0.00 0.62 0.01 32.42 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $730 $135 $542 $10 $2,724 $1,104 $19,908 $3,505 $4 $613 $5 $111,626 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results: SAMA 23 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,256,480 $15,231 2.0 $30,462
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-189 License Renewal Application E.6.23.2 EXTERNAL FLOODING EVALUATION This SAMA will have a limited impact for external flooding scenarios given that many of the scenarios result in flooding the plant's safety equipment, which would render the reactor building coolers inoperable. For cases in which power is available, there would be some benefit. The circumstances related to each flood range are discusses below:
* Floods over 310' msl:  For these floods, water level increases until it pours over the top of the existing flood barriers. Flooding of safety equipment occurs and the subsequent damage to the plant would result in a permanent loss of AC power. No credit is taken for this SAMA for this flood scenario.
* Floods between 305' and 310' msl:  Scenarios "A" through "D" are the result of flood gate failures. In these scenarios, the result is the same as for floods over 310' msl and no credit is taken for SAMA 23. In sequence "E", the flood gates hold, but EDG failures cause an SBO. This SAMA could be useful if power was recovered and there was a failure of the RR-V-6 valve to open. Given that only 0.7 percent of the internal events SBO contributors are "power recovered" cases (flooding cases are less likely to recover power due to extreme weather), that RR-V-6 failures contribute to less that 2.0 percent to the release frequency even when power is available, and that the total Sequence "E" frequency is only 3.66E-06, the impact of this SAMA would not be measurable. No credit is taken for this SAMA for sequence "E". For sequence "F", the operators have no warning of the impending flood and the flood gates are not installed in time to prevent flooding of the safety equipment. As with the other, similar sequences, no credit is taken for SAMA 23 for sequence "F".
* Floods below 305' mls:  These are similar to internal events LOOP scenarios and the recovery of RB cooling could be beneficial. The dose-risk and OECR of this flood sequence are assumed to be reduced in proportion to the internal events dose-risk and OECR. Implementation of this SAMA would result in only a limited risk reduction, as summarized below:  SAMA 23 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 8.11E-05 177.16 $542,152 Percent Change 0.0% 0.0% 0.0%
Environmental Report Appendix E  SAMA ANALYSIS Page E-190 Three Mile Island Nuclear Station Unit 1 License Renewal Application A further breakdown of this information is provided below according to release category. SAMA 23 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,212 $542,152 The corresponding external flooding component of the averted cost-risk is shown below: SAMA 23 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $15,543,306 $167 E.6.23.3 COST OF IMPLEMENTATION Procedure changes are estimated to be $50,000 (CPL 2004). E.6.23.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 23 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $30,462 $167 $30,629 $50,000 -$19,371 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-191 License Renewal Application E.6.24 SAMA NUMBER 24:  INSTALL DAMAGE RESISTANT HIGH TEMPERATURE RCP SEALS WITH A DIESEL ENGINE AS AN ALTERNATE DRIVE FOR AN EFW PUMP AND A PORTABLE 480V AC GENERATOR FOR EXTENDED EFW OPERATION For SBOs in which EFW has failed, neither primary nor secondary side cooling is available. Installing the enhanced RCP seals will prevent seal LOCAs and use of a portable generator would allow the turbine driven EFW pump to be used for extended periods in an SBO, as suggested in SAMA 2. However, in the event that the turbine driven EFW pump fails, there would be no means of providing secondary side makeup. Turbine driven EFW failures could be mitigated if an engine was available to drive one of the EFW pumps. Other industry SAMA applications have suggested similar strategies, but they typically suggest the turbine driven pumps as the best option for connection to the engine based on ease of connection. For scenarios with turbine driven EFW failure, however, the initial TD EFW pump failure may prevent its further use even with an alternate motive source. As a result, this SAMA, in addition to the requirements of SAMA 2, requires that the diesel engine be connected to one of the motor driven EFW pumps. E.6.24.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associated with the internal events and the non-external flooding events. As described in
 
Section E.4.6.3, the external events risk, excluding external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events based averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise the total averted cost-risk for a SAMA. To simulate the availability of a proposed alternate diesel-driven EFW pump, the cutsets from SAMA 2 were further adjusted by setting the turbine-driven EFW pump start and run failures to zero, i.e., EFEF-P-1----P7FS and EFEFP1------P7FR, respectively. Other contributors exist related to turbine driven EFW failures, but these are the major contributors and removing them from the cutsets is considered to adequately represent the benefits this SAMA. The new cutset probabilities for CDF and the various release categories were then summed and used to determine an estimate for the averted cost risk. Therefore, no new logic changes were made to the PRA model and no fault tree requantifications were performed. The following table summarizes the changes that were made to the basic event data:
Environmental Report Appendix E  SAMA ANALYSIS Page E-192 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 24 - Model Changes Gate and / or Basic Event ID and Description Description of Change EFEF-P-1----P7FS:  TURBINE-DRIVEN PUMP EF-P-1 FAILS TO START The basic event probability was changed from 4.66E-03 to 0.0. EFEFP1------P7FR: TURBINE-DRIVEN PUMP EF-P-1 FAILS
 
TO RU The basic event probability was changed from 5.06E-02 to 0.0. The model changes identified above yielded a reduction in the CDF, Dose-risk, and Offsite Economic cost-risk, as summarized below:  SAMA 24 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 1.11E-05 14.68 $54,017 Percent Change
-53.2% -55.0% -51.9% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 24 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.18E-07 8.46E-07 1.80E-07 9.29E-09 0.00E+000.00E+000.00E+000.00E+007.15E-091.25E-094.86E-11 2.67E-08 3.54E-074.88E-085.81E-09Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.39 4.84 0.91 0.05 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.08 2.18 0.30 0.02 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $11,620 $23,519 $3,348 $173 $0 $0 $0 $0 $64 $11 $0 $240 $7,151 $986 $55 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.43E-11 0.00E+00 5.24E-09 1.33E-09 6.66E-083.81E-105.57E-083.54E-086.34E-078.12E-064.86E-09 2.81E-07 7.97E-111.11E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.01 0.00 0.09 0.00 0.08 0.05 1.41 2.17 0.00 0.08 0.00 14.68 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $0 $0 $50 $13 $254 $1 $213 $135 $3,982 $2,127 $1 $74 $0 $54,017 Based on these results, the averted cost-risk for all non-external flooding contributors can be calculated using the 2.0 multiplier on the internal events results:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-193 License Renewal Application SAMA 24 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $1,536,137 $1,735,574 2.0 $3,471,148 E.6.24.2 EXTERNAL FLOODING EVALUATION This SAMA can have an impact on any SBO, seal LOCA, or EFW failure scenario. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, all safety equipment is flooded and this SAMA has no impact on the risk. No provisions are made for flood proofing the EFW pump in this SAMA. SAMA 32 addresses flood proof secondary side makeup capabilities.
* Floods between 305' and 310' msl:  Most of the sequences are not impacted by this SAMA as core damage is caused by failure of the flood gates (the safety equipment is flooded) or because a flood warning is not provided and no preparations are made for the flood (the safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of all AC power leads to core damage. These SBO cases are assumed to be completely mitigated by this SAMA.
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. As a result, these flooding sequences would be impacted in the same way as the internal events LOOP events. In order to simplify the calculations, SAMA 24 is assumed to eliminate all risk from this flooding sequence. The following tables summarize the results of these changes: SAMA 24 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 7.72E-05 166.08 $508,082 Percent Change
-4.8% -6.3% -6.3% A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E  SAMA ANALYSIS Page E-194 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 24 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 0.00E+00 8.65E-08 0.00E+00 7.72E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 0.00 0.25 0.00 166.08 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839
$0 $778 $0 $508,082 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 24 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $14,598,420 $945,053 E.6.24.3 COST OF IMPLEMENTATION The cost of implementation for this SAMA is estimated to be a combination of SAMA 2
($7,300,000) and the $1.1 million estimate for a direct drive diesel injection pump from Palisades (NMC 2005). The total implementation cost is $8,400,000. E.6.24.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 24 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $3,471,148 $945,053 $4,416,201 $8,400,000 -$3,983,799 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-195 License Renewal Application E.6.25 SAMA NUMBER 25:
INSTALL AN ADDITIONAL EDG An additional source of AC power is a potential means of supplying an entire division of safety equipment in the event that on-site AC power is lost in a LOOP. While additional EDGs are expensive, they can be cost effective at some plants, especially those with a large LOOP/SBO contribution to CDF. However, for TMI-1, the SBO EDG is available at the site and a less costly solution to reducing risk through AC power improvements would be to implement SAMA 1 rather than to install an additional EDG. Without auto alignment capability, the benefit of a new EDG would not be maximized and installing an additional EDG with auto alignment capability would be illogical when the existing SBO EDG could be upgraded first. Therefore, installation of an additional EDG would imply that SAMA 1 must already be installed. In this case, the benefit of installing an additional EDG is approximated assuming previous installation of SAMA 1, but the cost of SAMA 1 is not included in the SAMA 25 implementation cost. E.6.25.2 NON-EXTERNAL FLOODING EVALUATION Rather than perform a full scale model change to evaluate this SAMA, PRA insights from the SAMA 1 results can be used to show that adding an additional EDG after implementing SAMA 1 would not be cost effective. The RRW values for the SBO EDG "fail to start", "fail to run", and maintenance terms based on both CDF and Level 2 are provided in the table below from the SAMA 1 importance lists. These are the main contributors to SBO EDG failures:
BASIC EVENT CDF BASED RRW VALUE LEVEL 2 BASED RRW VALUE GSEG-Y-4----DGFS: STATION BLACKOUT DG FAILS TO START 1.010 1.025 GS-SBODG----DGFR: SBO DIESEL FAILS TO RUN 1.019 1.049 GS-EG-Y-4---DGMM: SBO Diesel Generator in Maintenance 1.011 1.029 Equivalent RRW Value of Events = 1.04 1.103 For independent events such as these, the RRW values can be combined to obtain a total RRW factor. Of the "equivalent" RRW values above, the Level 2 value is the larger of the two results and if the larger 1.103 RRW is assumed to apply to both the CDF and the Level 2 results, the impact of eliminating SBO EDG failures can be estimated. This is done by dividing the SAMA 1 internal events MACR of $5,580,172 by 1.103 and subtracting the result from the SAMA 1 internal events MACR, which yields $521,086 ($5,580,172 - ($5,580,172 / 1.103) = $521,086).
Environmental Report Appendix E  SAMA ANALYSIS Page E-196 Three Mile Island Nuclear Station Unit 1 License Renewal Application This can be done because the relationship between the MACR and the frequencies is linear and because the larger of the two "equivalent" RRW values was used to represent both the Level 1 and Level 2 results. E.6.25.3 EXTERNAL FLOODING EVALUATION This SAMA can have an impact on any scenario requiring the operation of the SBO EDG. For the external flooding cases, the three flood regimes are impacted differently:
* Floods over 310' msl:  In these scenarios, the safety equipment is flooded and the addition of another EDG would have no impact on risk.
* Floods between 305' and 310' msl:  Most of the sequences would not be impacted by the addition of another EDG as core damage is caused by failure of the flood gates (the safety equipment is flooded) or because a flood warning is not provided and no preparations are made for the flood (the safety equipment is flooded). Flood sequence "E" represents cases where the flood gates are correctly installed, but a loss of all AC power leads to core damage. In these cases, the installation of another EDG could provide a large benefit. For the purposes of this evaluation, it is assumed that SAMA 25 will eliminate all of the contribution from Sequence "E".
* Floods below 305' mls:  The only impact these flood scenarios have on the plant is the potential to cause a loss of offsite power. SAMA 25 would have an impact on these SAMAs and for the purposes of this analysis, all risk from this flood sequence is assumed to be eliminated. The following tables summarize the results of these changes: SAMA 25 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 7.72E-05 166.08 $508,082 Percent Change
-4.8% -6.3% -6.3% A further breakdown of this information is provided below according to release category.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-197 License Renewal Application SAMA 25 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 0.00E+00 8.65E-08 0.00E+00 7.72E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 132.75 13.13 0.19 1.89 17.87 0.00 0.25 0.00 166.08 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $405,951 $40,149 $598 $5,767 $54,839
$0 $778 $0 $508,082 Based on these results, the external flooding component of the averted cost-risk can then be calculated: SAMA 25 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $14,598,420 $945,053 E.6.25.3 COST OF IMPLEMENTATION Browns Ferry estimated the cost of installing an additional EDG to be $6 million. While there are estimates as high as $25 million used in SAMA analyses for the installation of additional EDGs, the Browns Ferry estimate is used for TMI-1. E.6.25.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 25 - Net Value Non-External Flooding Based Averted Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $521,086 $945,053 $1,466,139 $6,000,000 -$4,533,861 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E  SAMA ANALYSIS Page E-198 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.26 SAMA NUMBER 26:  REROU T E CAB L ES SO THAT THEY DO NOT PASS OVER IGNITION SOURCES IN FIRE ZONE CB-FA-2E (WEST INVERTER ROOM) OR WRAP THEM IN FIRE PROOF MATERIAL The TMI-1 IPEEE fire analysis identified that cables important to control functions for essential AC panels (including Bus 1E) were routed over potential ignition sources in fire zone CB-FA-2E. Some of the risk from this fire zone could be averted if these cables were protected or rerouted such that battery charger/inverter fires would not result in damage to the cables. While these changes would not eliminate the risk corresponding to the ignition source fires, the cables are the dominant risk contributors for the zone. Two potential methods of mitigating the fire risk in CB-FA-2E have been identified for this SAMA
* Method A:  Rerouting the cables so that they do not pass over the battery chargers or inverters or,
* Method B:  Providing fire barriers capable of preventing fire propagation and damage to the overhead cables. Both of these changes are assumed to be capable of preventing damage to the overhead cables. Rerouting the cables has the potential to completely eliminate the risk of cable damage while use of fire barriers has some non-zero failure rate associated with the barriers, but for this analysis, both approaches are assumed to completely eliminate the risk of cable damage. The impact of these types of changes has been es timated using available information from the fire model and engineering judgment. No model quantification was performed for this evaluation. It is assumed that if the portion of the fire CDF and release consequences related to cable damage in fire zone CB-FA-2E can be identified, then an averted cost-risk can be calculated for this SAMA. The steps used to perform this calculation are provided below:
* Determine the component of the overall modified MACR attributable to non-external flooding external events,
* Determine the component of the non-external flooding external events cost-risk attributable to fire events,
* Determine the component of the fire based cost-risk attributable to fire zone CB-FA-2E, Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-199 License Renewal Application
* Determine the component of the fire based cost-risk attributable to the AC panel control cables located in fire zone CB-FA-2E,
* Calculate the percent reduction in the fire CB-FA-2E CDF that would occur if the SAMA is implemented and reduce the cost-risk for the fire zone by the same percent. The reduction in cost-risk is the averted cost-risk for this SAMA. The baseline assumption for non-external flooding external events contributions in the TMI-1 SAMA is that they are approximately equal to the internal events contributions. Given that the internal events MACR is $3,271,711, the same value is assigned to external events. The relative contribution of fire events to the total external events CDF is difficult to determine due to the fact that the methods of analysis for each of the external events types are not necessarily compatible. If the comparison is made strictly on the basis of the calculated CDFs, the fire contribution would only be 20.1%: External Events Contribution Summary External Event CDF Percent of Total Non-External Flooding External Events CDF Seismic (based on LLNL seismic hazard curves) 8.43E-05/yr 78.6% Fire* 2.16E-05/yr 20.1% High Winds 7.77E-07/yr 0.7% Aircraft Impact** 3.95E-07/yr 0.4% Hazardous Chemicals 1.60E-07/yr 0.1% Total 1.07E-04/yr  *Includes the error in the IPEEE that results in overestimation of the CB-FA-2E fire zone frequency. **This includes the contribution from accidental aircraft impact only. Intentional aircraft impact is addressed in separate plant programs and is beyond the scope of the SAMA analysis.
For seismically stable regions, the fire CDF is typically greater than the seismic CDF and for TMI-1, a larger value than the 20 percent identified in the table above is considered to be appropriate. For the purposes of this calculation, the fire CDF is assumed to be 85 percent of the total non-external flooding external events CDF. This corresponds to a cost-risk of
$2,780,954 ($3,271,711
* 0.85 = $2,780,954).
Environmental Report Appendix E  SAMA ANALYSIS Page E-200 Three Mile Island Nuclear Station Unit 1 License Renewal Application The cost-risk associated with fire zone CB-FA-2E can then be determined based on its relative contribution to the total fire CDF by assuming the fire zone specific MACR is directly proportional to the CDF. For this calculation, the error identified in the IPEEE related to the CB-FA-2E CDF has been corrected:
Fire Area/Scenario Description CDF 1 Percent of Total Fire CDF Fire Zone Specific MACR CB-FA-2d East Inverter Room 4.94E-06/yr 26.17% $727,776 CB-FA-2e West Inverter Room 3.09E-06/yr 16.31% $453,574 CB-FA-3a 1D Switchgear Room 3.94E-06/yr 20.87% $580,385 CB-FA-3b 1E Switchgear Room 4.96E-06/yr 26.27% $730,557 CB-FA-4b Control Room - Console CR 1.96E-06/yr 10.38% $288,663  Total1.89E-05/yr 100% $2,780,955 The risk reduction possible for fire zone CB-FA-2E is a fraction of the total based on the potential capabilities of the changes proposed in this SAMA. Neither change (barriers or cable rerouting) is considered to be capable of preventing damage to the equipment in the cabinet where the fire starts; however, both changes are assumed to prevent damage to overhead cables. The averted cost-risk for these changes, therefore, is based on the difference between the CDF when cable damage occurs and the CDF when cable damage is eliminated. The quantification of the CDF change due to this SAMA's implementation was performed using information from the IPEEE documentation. The IPEEE indicates that the CDF for fire zone CB-FA-2E is composed of two cases that are separated based on the location of the two batteries and two inverters in the zone. One battery and one inverter are located below the AC panel control cables and the other battery and inverter are not located below the AC power control cables. Fires in the battery or inverter below the AC control cables damage essential AC power (and are assumed to fail bus 1E), but fires in the battery or inverter located away from the AC panel control cables do not damage the AC control panel cables. The fire zone CDF for the existing plant configuration is summarized in the following table:
 
1 The CB-2A-FE fire zone CDF reported in the IPEEE appears to have been overestimated due to computational errors. The correct CDF calculation for fire zone CB-FA-2E is shown here and used in the remainder of this calculation.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-201 License Renewal Application Current CB-FA-2E Fire Contributions  Conditional CDF IE Frequency Fraction of  IE Frequency Applicable CDF Case 1 (fires not resulting in cable failures) 1.16E-04 4.91E-03 5.00E-01 2.85E-07 Case 2 (fires resulting in cable failures) 1.14E-03 4.91E-03 5.00E-01 2.80E-06    Total 3.08E-06 To represent implementation of the SAMA, Case 2 is adjusted such that the conditional CDF is equal to the conditional CDF for Case 1, which implies that the AC control cables are not damaged and that the consequences of failing either battery/inverter set are the same. The
 
CDF results are shown below: POST SAMA CB-FA-2E FIRE CONTRIBUTIONS  CONDITIONAL CDF IE FREQUENCY FRACTION OF IE FREQUENCY APPLICABLE CDF Case 1 (fires not resulting in cable failures) 1.16E-04 4.91E-03 5.00E-01 2.85E-07 Case 2 (fires resulting in cable failures) 1.16E-04 4.91E-03 5.00E-01 2.85E-07 Total 5.70E-07 The result is a CDF of 5.70E-7, which is 18.5 percent of the base CB-FA-2E CDF. This corresponds to a revised fire zone MACR of $83,911 ($453,574
* 0.185 = $83,911). The difference between the baseline MACR for fire zone CB-FA-2E and MACR assuming SAMA implementation is the averted cost-risk for this SAMA:  $453,574 - $83,911 = $369,663. Of the two potential mitigation methods identified, cable wrapping (Method B) was determined to be the more cost effective approach. The cost of performing the cable wrapping in CB-FA-2e was estimated to be $900,000 by the TMI staff (Exelon 2007c).
Results The results of the fire area analysis and the implementation cost estimates are used as input to the cost-benefit calculation. The results of this calculation are provided in the following table:
Environmental Report Appendix E  SAMA ANALYSIS Page E-202 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 26 - Net Value Averted Cost-Risk Cost of Implementation Net Value $369,663 $900,000 -$530,337 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
E.6.27 SAMA NUMBER 27:  IMPR O VE THE 480V AC LOAD CEN T ER WELDS The IPEEE determined that the existing 480V AC load centers had the lowest seismic fragilities in the TMI-1 AC distribution system. Adding reinforcements to the welds on the load center framework would improve the seismic durability of the structure and increase the likelihood that the system would be available after a seismic event. The specific components considered to be addressed are 480V AC load centers 1P, 1R, 1S, and 1T, which are the components critical to improving the AC power system's seismic ruggedness. The other low seismic capacity components of the AC distribution system, the EDG air receivers, were enhanced subsequent to
 
the completion of the IPEEE. The ability to quantify the impact of improving the seismic capacity of the load centers is limited due to the small amount of information provided in the IPEEE related to the importance of the load centers over the four different seismic ranges evaluated. However, a process has been developed to approximate the potential benefit of increasing the HCLPF for the load centers from 0.12g to 0.30g through improvements to the welds. The revised HCLPF capacity value of 0.30g was chosen because it was used in industry seismic margins analyses as the threshold for components to be considered adequately durable. All of the calculations are based on information available in the IPEEE, the current PRA, and engineering judgment. No seismic model quantification was performed for this evaluation. It is assumed that if the portion of the seismic CDF and release consequences related to the failures of the 480V AC load centers can be identified, then an averted cost-risk can be calculated for this SAMA. The steps used to perform this calculation are provided below:
* Determine the component of the overall modified MACR attributable to non-external flooding external events,
* Determine the component of the non-external flooding external events cost-risk attributable Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-203 License Renewal Application to seismic events,
* Determine the component of the seismic based cost-risk attributable to 480V AC load centers 1P, 1R, 1S, and 1T,
* Calculate the percent reduction in seismic CDF that would occur if the SAMA is implemented and reduce the cost-risk for the load centers by the same percent. The reduction in cost-risk is the averted cost-risk for this SAMA. The baseline assumption for non-external flooding external events contributions in the TMI-1 SAMA is that they are approximately equal to the internal events contributions. Given that the internal events MACR is $3,271,711, the same value is assigned to external events. The relative contribution of seismic events to the total external events CDF is difficult to determine due to the fact that the methods of analysis for each of the external events types are not necessarily compatible. If the comparison is made strictly on the basis of the calculated CDFs, the seismic contribution would be 78.6%:
External Events Contribution Summary External Event CDF Percent of Total  Non-External Flooding External Events CDF Seismic (based on LLNL seismic hazard curves) 8.43E-05/yr 78.6% Fire* 2.16E-05/yr 20.1% High Winds 7.77E-07/yr 0.7% Aircraft Impact** 3.95E-07/yr 0.4% Hazardous Chemicals 1.60E-07/yr 0.1% Total 1.07E-04/yr  *Includes the error in the IPEEE that results in overestimation of the CB-FA-2E fire zone frequency. **This includes the contribution from accidental aircraft impact only. Intentional aircraft impact is addressed in separate plant programs and is beyond the scope of the SAMA analysis.
For seismically stable regions, the fire CDF is typically greater than the seismic CDF, but for TMI-1, this is not the case when the NUREG 1488 LLNL hazard curves are used. While it may Environmental Report Appendix E  SAMA ANALYSIS Page E-204 Three Mile Island Nuclear Station Unit 1 License Renewal Application be inconsistent with many industry examples in which the fire risk outweighs the seismic risk, the 78.6 percent seismic contribution is retained for this evaluation. This corresponds to a cost-risk of $2,571,565 ($3,271,711
* 0.786 = $2,571,565). The cost-risk associated with the 480V AC load centers can then be determined based on the overall seismic Fussell-Vesely (F-V) value for the load centers and the assumption that the overall seismic F-V value is constant over the seismic spectrum. This is typically not true, but when used over the entire seismic spectrum, it will provide a reasonable answer. Two separate F-V values have been identified for the 480V AC load centers, which are part of the FRAG15 component group (based on the NUREG-1488 seismic hazard curve results):
* GW: offsite power available cases (F-V = 0.4),
* GY: offsite power failure cases (F-V = 0.15). The CDF corresponding to the FRAG15 component group (the 480V load centers) can be estimated by multiplying the F-V values by the CDF for each range in the seismic spectrum, as summarized below: GW FRAG15 Specific CDF Initiating Event CDF CDF Related to FRAG15 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.26E-05 5.04E-06 SEIS2 (0.25g) (range = 0.2g - 0.3g) 2.61E-05 1.04E-05 SEIS3 (0.4g) (range = 0.3g - 0.5g) 3.25E-05 1.30E-05 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.31E-05 5.24E-06 Totals= 8.43E-05 3.37E-05  GY FRAG15 Specific CDF Initiating Event CDF CDF Related to FRAG15 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.26E-05 1.89E-06 SEIS2 (0.25g) (range = 0.2g - 0.3g) 2.61E-05 3.92E-06 SEIS3 (0.4g) (range = 0.3g - 0.5g) 3.25E-05 4.88E-06 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.31E-05 1.97E-06 Totals= 8.43E-05 1.26E-05 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-205 License Renewal Application Assuming the MACR is directly proportional to the CDF provides a means of determining the MACR for FRAG15 over the seismic spectrum given the total seismic MACR of $2,571,565: GW FRAG15 Specific MACR Initiating Event CDF Related to FRAG15 MACR Related to FRAG15 SEIS1 (0.15g) (range = 0.052g - 0.2g) 5.04E-06 $153,745 SEIS2 (0.25g) (range = 0.2g - 0.3g) 1.04E-05 $318,471 SEIS3 (0.4g) (range = 0.3g - 0.5g) 1.30E-05 $396,564 SEIS4 (0.6g) (range = 0.5g - 1.01g) 5.24E-06 $159,846 Totals= 3.37E-05 $1,028,626 GY FRAG15 SPECIFIC MACR INITIATING EVENT CDF RELATED TO FRAG15 MACR RELATED TO FRAG15 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.89E-06 $57,654 SEIS2 (0.25g) (range = 0.2g - 0.3g) 3.92E-06 $119,427 SEIS3 (0.4g) (range = 0.3g - 0.5g) 4.88E-06 $148,711 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.97E-06 $59,942 Totals= 1.26E-05 $385,735  The quantification of the CDF change due to this SAMA's implementation was performed using information from the IPEEE documentation. The IPEEE indicates that the HCLPF capacity for FRAG15 is 0.12g and the failure probabilities for each seismic range are explicitly provided for FRAG15. In addition, the failure probabilities ar e explicitly provided for the BWST (FRAG21),
which has a HCLPF capacity of 0.3g. It is assumed that if the 480V AC load center welds are improved, the failure probabilities can be repr esented by those documented for FRAG21. From these assumptions the revised CDFs, and therefore the MACRs, can be calculated. More specifically, the ratio of the post-SAMA FRAG15 failure probability to the baseline FRAG15 failure probability will be equivalent to the ratio of the post-SAMA FRAG15 CDF to the baseline FRAG15 CDF. Finally, the FRAG15 MACR is proportional to the CDF, so once the FRAG15 CDF ratio is known, the post-SAMA FRAG15 MACR can be calculated by multiplying the FRAG15 ratio by the baseline FRAG15 MACR for each seismic hazard range. The following tables summarize the results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-206 Three Mile Island Nuclear Station Unit 1 License Renewal Application GW FRAG15 Specific MACR Post SAMA Implementation Initiating Event Baseline FRAG15 Failure Probability Post-SAMA FRAG15 Failure Probability (0.3g HCLPF)
CDF (or FRAG15) Ratio Baseline FRAG15 MACR  Post-SAMA FRAG15 MACR (0.3g HCLPF) SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.25E-02 2.15E-06 1.72E-04 $153,745 $26 SEIS2 (0.25g) (range =
0.2g - 0.3g) 2.67E-01 3.95E-03 1.48E-02 $318,471 $4,711 SEIS3 (0.4g) (range =
0.3g - 0.5g) 6.61E-01 4.78E-02 7.23E-02 $396,564 $28,677 SEIS4 (0.6g) (range =  0.5g - 1.01g) 9.50E-01 2.82E-01 2.97E-01 $159,846 $47,449 Total = $1,028,626 $80,864  GY FRAG15 SPECIFIC MACR POST SAMA IMPLEMENTATION Initiating Event Baseline FRAG15 Failure Probability Post-SAMA FRAG15 Failure Probability (0.3g HCLPF)CDF Ratio Baseline FRAG15 MACR  Post-SAMA FRAG15 MACR (0.3g HCLPF) SEIS1 (0.15g) (range =  0.052g - 0.2g) 1.58E-02 2.15E-06 1.36E-04 $57,654 $8 SEIS2 (0.25g) (range =
0.2g - 0.3g) 3.60E-01 3.95E-03 1.10E-02 $119,427 $1,310 SEIS3 (0.4g) (range =
0.3g - 0.5g) 8.44E-01 4.78E-02 5.66E-02 $148,711 $8,422 SEIS4 (0.6g) (range =  0.5g - 1.01g) 9.98E-01 2.82E-01 2.83E-01 $59,942 $16,938 Total = $385,735 $26,678 The averted cost-risk is the difference bet ween the base FRAG15 MACRs and the post-SAMA implementation MACRs for both GW and GY, which is $1,306,819 (($1,028,626 - $80,864) +
($385,735 - $26,678) = $1,306,819). E.6.27.1 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $575,000 by the TMI staff (Exelon 2007c).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-207 License Renewal Application E.6.27.2 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is only the seismic averted cost-risk in this case, and the cost of implementation. The following table summarizes these results: SAMA 27 - Net Value Total Averted Cost-Risk Cost of Implementation Net Value $1,306,819 $575,000 $731,819 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive.
E.6.28 SAMA NUMBER 28:  IMPROVE THE DECAY HEAT SERVICE COOLER (DC-C-2A/B) ANCHORAGES The IPEEE determined that the existing Decay heat service coolers (DC-C-2A/B) lacked sufficiently durable anchorages. Replacing the anchorages with more robust anchorages would improve the seismic durability of the structure and increase the likelihood that the heat exchangers would be available after a seismic event. The ability to quantify the impact of improving the seismic capacity of the heat exchanger anchorages is limited due to the small amount of information provided in the IPEEE related to the importance of DC-C-2A/B over the four different seismic ranges evaluated. However, a process has been developed to approximate the potential benefit of increasing the HCLPF for the heat exchangers from 0.09g to 0.30g through improvements to the anchorages. The revised HCLPF capacity value of 0.30g was chosen because it was used in industry seismic margins analyses as the threshold for components to be considered adequately durable. All of the calculations are based on information available in the IPEEE, the current PRA, and engineering judgment. No seismic model quantification was performed for this evaluation. It is assumed that if the portion of the seismic CDF and release consequences related to the failures of DC-C-2A/B can be identified, then an averted cost-risk can be calculated for this SAMA. The steps used to perform this calculation are provided below:
* Determine the component of the overall modified MACR attributable to non-external flooding external events, Environmental Report Appendix E  SAMA ANALYSIS Page E-208 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Determine the component of the non-external flooding external events cost-risk attributable to seismic events,
* Determine the component of the seismic based cost-risk attributable to Decay Heat Service Coolers DC-C-2A/B,
* Calculate the percent reduction in seismic CDF that would occur if the SAMA is implemented and reduce the cost-risk for the heat exchangers by the same percent. The reduction in cost-risk is the averted cost-risk for this SAMA. The baseline assumption for non-external flooding external events contributions in the TMI-1 SAMA is that they are approximately equal to the internal events contributions. Given that the internal events MACR is $3,271,711, the same value is assigned to external events. The relative contribution of seismic events to the total external events CDF is difficult to determine due to the fact that the methods of analysis for each of the external events types are not necessarily compatible. If the comparison is made strictly on the basis of the calculated CDFs, the seismic contribution would be 78.6%: External Events Contribution Summary External Event CDF Percent of Total Non-External Flooding External Events CDF Seismic (based on LLNL seismic hazard curves) 8.43E-05/yr 78.6% Fire* 2.16E-05/yr 20.1% High Winds 7.77E-07/yr 0.7% Aircraft Impact** 3.95E-07/yr 0.4% Hazardous Chemicals 1.60E-07/yr 0.1% Total 1.07E-04/yr  *Includes the error in the IPEEE that results in overestimation of the CB-FA-2E fire zone frequency. **This includes the contribution from accidental aircraft impact only. Intentional aircraft impact is addressed in separate plant programs and is beyond the scope of the SAMA analysis.
For seismically stable regions, the fire CDF is typically greater than the seismic CDF, but for TMI-1, this is not the case when the NUREG 1488 LLNL hazard curves are used. While it may Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-209 License Renewal Application be inconsistent with many industry examples in which the fire risk outweighs the seismic risk, the 78.6 percent seismic contribution is retained for this evaluation. This corresponds to a cost-risk of $2,571,565 ($3,271,711
* 0.786 = $2,571,565). The cost-risk associated with DC-C-2A/B can then be determined based on the overall seismic Fussell-Vesely (F-V) value for the heat exchangers and the assumption that the overall seismic F-V value is constant over the seismic spectrum. This is typically not true, but when used over the entire seismic spectrum, it will provide a reasonable answer. The overall seismic F-V value for component group FRAG11, which includes DC-C-2A/B, is 2.00E-02 (based on the NUREG-1488 seismic hazard curve results). The CDF corresponding to the FRAG11 component group (the Decay Heat Service Coolers) can be estimated by multiplying the FRAG11 F-V value by the CDF for each range in the seismic spectrum. The following table summarizes the results:
FRAG11 Specific CDF Initiating Event CDF CDF Related to FRAG11 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.26E-05 2.52E-07 SEIS2 (0.25g) (range = 0.2g - 0.3g) 2.61E-05 5.22E-07 SEIS3 (0.4g) (range = 0.3g - 0.5g) 3.25E-05 6.50E-07 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.31E-05 2.62E-07 Totals= 8.43E-05 1.69E-06 Assuming the MACR is directly proportional to the CDF provides a means of determining the MACR for FRAG11 over the seismic spectrum given the total seismic MACR of $2,571,565:
FRAG11 Specific MACR Initiating Event CDF Related to FRAG11 MACR Related to FRAG11 SEIS1 (0.15g) (range = 0.052g - 0.2g) 2.52E-07 $7,687 SEIS2 (0.25g) (range = 0.2g - 0.3g) 5.22E-07 $15,924 SEIS3 (0.4g) (range = 0.3g - 0.5g) 6.50E-07 $19,828 SEIS4 (0.6g) (range = 0.5g - 1.01g) 2.62E-07 $7,992 Totals= 1.69E-06 $51,431 The quantification of the CDF change due to this SAMA's implementation was performed using information from the IPEEE documentation. The IPEEE provides the seismic range specific failure probabilities for top event RX, which are driven by FRAG11 given that the HCLPF Environmental Report Appendix E  SAMA ANALYSIS Page E-210 Three Mile Island Nuclear Station Unit 1 License Renewal Application capacity is 0.09g while the only other contributing component has a HCLPF capacity of 0.43g. In addition, the failure probabilities are explicitly provided for the BWST (FRAG21), which has a HCLPF capacity of 0.30g. It is assumed that if the DC-C-2A/B anchorages are improved, the failure probabilities can be represented by those documented for FRAG21 (HCLPF for DC-C-2A/B improved to 0.30g). From these assumptions, the revised CDFs and the corresponding MACRs can be calculated using the ratio of the revised CDFs to the original CDFs. The following tables summarize the results: FRAG11 Specific MACR Post SAMA Implementation Initiating Event Base FRAG11 Failure Probability (From top event RX) FRAG11 Failure Probability After SAMA Implementation (0.3g HCLPF) CDF Ratio Baseline FRAG11 MACR Post-SAMA FRAG11 MACR (0.3g HCLPF) SEIS1 (0.15g) (range = 0.052g - 0.2g) 3.46E-02 2.15E-06 6.21E-05 $7,687
$0 SEIS2 (0.25g) (range = 0.2g - 0.3g) 4.82E-01 3.95E-03 8.20E-03 $15,924 $130 SEIS3 (0.4g) (range = 0.3g - 0.5g) 8.42E-01 4.78E-02 5.68E-02 $19,828 $1,126 SEIS4 (0.6g) (range = 0.5g - 1.01g) 9.87E-01 2.82E-01 2.86E-01 $7,992 $2,284 Total = $51,431 $3,540 The averted cost-risk is the difference between the base FRAG11 specific MACRs and the FRAG11 specific MACRS after SAMA implementation, which is $47,891 ($51,431 - $3,540 =
 
$47,891). E.6.28.1 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $575,000 by the TMI staff (Exelon 2007c). E.6.28.2 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is only the seismic averted cost-risk in this case, and the cost of implementation. The following table summarizes these results: SAMA 28 - Net Value Total Averted Cost-Risk Cost of Implementation Net Value $47,891 $575,000 -$527,109 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-211 License Renewal Application E.6.29 SAMA NUMBER 29:  REPLACE EDG GROUND RES I STORS Failure of the EDG ground resistors results in failure of the EDGs, which will lead to core damage in the event that off-site power is not available. Given that the HCLPF capacity for these components was estimated at 0.25g compared with 0.09g capacities of off-site power components (such as the 1/A and 1/B distribution buses or the aux transformers), it is likely that core damage will ensue due to long term loss of power if the EDG ground resistors fail from seismic shock. Replacing the resistors with more durable versions would improve the reliability of the EDGs in seismic events. The ability to quantify the impact of improving the seismic capacity of the EDG ground resistors is limited due to the small amount of information provided in the IPEEE related to the importance of these components over the four di fferent seismic ranges evaluated. However, a process has been developed to approximate the potential benefit of increasing the HCLPF capacity of the EDG ground resistors from 0.25g to a theoretical limit where it would never fail.
All of the calculations are based on information available in the IPEEE, the current PRA, and engineering judgment. No seismic model quantification was performed for this evaluation. It is assumed that if the portion of the seismic CDF and release consequences related to the failures of the EDG ground resistors can be identified, then an averted cost-risk can be calculated for this SAMA. The steps used to perform this calculation are provided below:
* Determine the component of the overall modified MACR attributable to non-external flooding external events,
* Determine the component of the non-external flooding external events cost-risk attributable to seismic events,
* Determine the component of the seismic based cost-risk attributable to the EDG ground resistors,
* Assume that implementation of this SAMA would eliminate all risk related to the EDG ground resistors such that the averted cost-risk would be the total cost-risk related to the EDG ground resistors.
Environmental Report Appendix E  SAMA ANALYSIS Page E-212 Three Mile Island Nuclear Station Unit 1 License Renewal Application The baseline assumption for non-external flooding external events contributions in the TMI-1 SAMA is that they are approximately equal to the internal events contributions. Given that the internal events MACR is $3,271,711, the same value is assigned to external events. The relative contribution of seismic events to the total external events CDF is difficult to determine due to the fact that the methods of analysis for each of the external events types are not necessarily compatible. If the comparison is made strictly on the basis of the calculated CDFs, the seismic contribution would be 78.6%: External Events Contribution Summary External Event CDF Percent of Total Non-External Flooding External Events CDF Seismic (based on LLNL seismic hazard curves) 8.43E-05/yr 78.6% Fire* 2.16E-05/yr 20.1% High Winds 7.77E-07/yr 0.7% Aircraft Impact** 3.95E-07/yr 0.4% Hazardous Chemicals 1.60E-07/yr 0.1% Total 1.07E-04/yr  *Includes the error in the IPEEE that results in overestimation of the CB-FA-2E fire zone frequency. **This includes the contribution from accidental aircraft impact only. Intentional aircraft impact is addressed in separate plant programs and is beyond the scope of the SAMA analysis.
For seismically stable regions, the fire CDF is typically greater than the seismic CDF, but for TMI-1, this is not the case when the NUREG 1488 LLNL hazard curves are used. While it may be inconsistent with many industry examples in which the fire risk outweighs the seismic risk, the 78.6 percent seismic contribution is retained for this evaluation. This corresponds to a cost-risk of $2,571,565 ($3,271,711
* 0.786 = $2,571,565). The cost-risk associated with the EDG ground resistors can then be determined based on the overall seismic Fussell-Vesely (F-V) value for the EDG ground resistors and the assumption that the overall seismic F-V value is constant over the seismic spectrum. This is typically not true, but when used over the entire seismic spectrum, it will provide a reasonable answer. The Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-213 License Renewal Application overall seismic F-V value for FRAG17, which represents the EDG ground resistors, is 1.00E-02 (based on the NUREG-1488 seismic hazard curve results). The CDF corresponding to the FRAG17 component group (the EDG ground resistors) can be determined by multiplying the FRAG-17 F-V value by the CDF for each range in the seismic spectrum. The following table summarizes the results:
FRAG17 Specific CDF Initiating Event CDF CDF Related to FRAG17 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.26E-05 1.26E-07 SEIS2 (0.25g) (range = 0.2g - 0.3g) 2.61E-05 2.61E-07 SEIS3 (0.4g) (range = 0.3g - 0.5g) 3.25E-05 3.25E-07 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.31E-05 1.31E-07 Totals= 8.43E-05 8.43E-07 Assuming the MACR is directly proportional to the CDF provides a means of determining the MACR for FRAG17 over the seismic spectrum given the total seismic MACR of $2,571,565:
FRAG17 Specific MACR Initiating Event CDF Related to FRAG17 MACR Related to FRAG17 SEIS1 (0.15g) (range = 0.052g - 0.2g) 1.26E-07
$3,844 SEIS2 (0.25g) (range = 0.2g - 0.3g) 2.61E-07 $7,962 SEIS3 (0.4g) (range = 0.3g - 0.5g) 3.25E-07
$9,914 SEIS4 (0.6g) (range = 0.5g - 1.01g) 1.31E-07
$3,996 Totals= 8.43E-07 $25,716 Following the assumption that implementation of this SAMA can eliminate all risk related to the EDG ground resistors, the averted cost-risk is the total MACR for FRAG17, which is $25,716. E.6.29.1 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $800,000 by the TMI staff (Exelon 2007c). E.6.29.2 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is only the seismic averted cost-risk in this case, and the cost of implementation. The following table summarizes these results:
Environmental Report Appendix E  SAMA ANALYSIS Page E-214 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 29 - Net Value Total Averted Cost-Risk Cost of Implementation Net Value $25,716 $800,000 -$774,284 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative. E.6.30 SAMA NUMBER 30:  IMPROVE DI ESEL FIRE PUMP FUEL OIL TANK AND BATTERY RACK SUPPORTS The Fire Service Water system provides cooling to the SBO EDG, backup cooling the DHCCW heat exchangers, and backup cooling to the "1A" and "1B" Instrument Air compressors. While seismic failures to the systems FSW supports would likely limit the benefit of improving the fuel oil tank and battery racks, some benefit may be available through improvements to the diesel fire pump's reliability. The ability to quantify the impact of improving the seismic capacity of the diesel fire pump is limited due to the small amount of information provided in the IPEEE related to the importance of the fire system. The motor driven pump (FS-P-2) appears to be explicitly included in the mode, but the diesel driven pumps (FS-P-1, FS-P-3) are not. However, a process has been developed to approximate the potential benefit of increasing the HCLPF capacity of the diesel driven pumps to a theoretical limit where they would never fail based on the seismic F-V value of the lowest contributor. All of the calculations are based on information available in the IPEEE, the current PRA, and engineering judgment. No seismic model quantification was performed for this evaluation. It is assumed that if the portion of the seismic CDF and release consequences related to the failures of the diesel driven fire pump can be identified, then an averted cost-risk can be calculated for this SAMA. The steps used to perform this calculation are provided below:
* Determine the component of the overall modified MACR attributable to non-external flooding external events,
* Determine the component of the non-external flooding external events cost-risk attributable to seismic events,
* Determine the component of the seismic based cost-risk attributable to the lowest reported Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-215 License Renewal Application seismic component group (EDG ground resistors),
* Assume that the seismic importance of the diesel driven fire pumps is equivalent to the EDG ground resistors,
* Assume that implementation of this SAMA would eliminate all risk related to the diesel driven fire pumps such that the averted cost-risk would be the total cost-risk related to the diesel driven fire pumps (equivalent to the MACR for the EDG ground resistors). Because neither the fire water system nor any fire water component was included in the seismic "system" or "component" importance lists, it is assumed that the MACR for the diesel fire driven pumps could not exceed that of the lowest component on the importance list. The IPEEE indicates that the lowest seismic F-V contributor is FRAG17, which represents the EDG ground
 
resistors.
Given that the MACR for FRAG17 was calcula t ed in Section E.6.29 , the calculations are not reproduced here, but the result was determined to be $25,716. This is considered to be the MACR for the diesel driven fire pumps. E.6.30.1 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $150,000 by the TMI staff (Exelon 2007c). E.6.30.2 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is only the seismic averted cost-risk in this case, and the cost of implementation. The following table summarizes these results: SAMA Number 30 Net Value Total Averted Cost-Risk Cost of Implementation Net Value $25,716 $150,000 -$124,284 Given that the cost of implementation is greater than the averted cost-risk for this SAMA, the net value is negative.
E.6.31 SAMA NUMBER 31:  MODIFY SPECIFIC CONTAINMENT P E N E TRATION MOVS TO FAIL CLOSED Most containment penetrations have AOV or SOV isolation valves that will fail closed on loss of air or power; however, there are cases in which MOVs are used instead. Those lines that do Environmental Report Appendix E  SAMA ANALYSIS Page E-216 Three Mile Island Nuclear Station Unit 1 License Renewal Application not include a pair of AOVs or SOVs that fail clos ed are typically below 1" in diameter or include at least one AOV or SOV that will fail closed on loss of air or power. However, the Nuclear Services Closed Cooling Water (NSCCW) and Reactor Building Normal Cooling (RBNC) systems include penetrations that only include MOVs:
* Valves NS-V-4, NS-V-15, NS-V-35 (NSCCW),
* Valves RB-V-2A, RB-V-7 (RBNC) While these are closed cooling systems that woul d not normally provide a credible release path, heat exchanger breaks in seismic events could provide containment bypass routes given that a break occurs in the reactor building as well. Changing one of the valves in each of these paths to fail closed is a means of increasing the isolation probability over what is available from manual action. Further review of the seismic design of the NSCCW and RBNC systems showed that while the heat exchangers linked to the penetrations in question were relatively weak, the piping and equipment associated with these lines within the reactor building were screened in the IPEEE as high capacity components. This indicates that failure of the piping and components within the reactor building would not occur except under the most extreme seismic conditions. In those cases, other integrity issues would likely exist and preventing a break in the NSCCW and RBNC lines would not provide any benefit. For reference purposes, an estimate of the cost required to replace the existing isolation valves with "fail closed", solenoid operated AOVs was prepared and determined to be $4,100,000 (Exelon 2007c), which is greater than the entire baseline external events cost-risk of $3,271,711. This SAMA is screened from further consideration. E.6.32 SAMA NUMBER 32:  PRE-STAGE SEVERE EXTERNAL FLOODING EQUIPMENT The existing severe flooding guidelines, which address external floods greater than 309' msl (stillwater level, 310' msl assumed wave action level), provide the TSC with information and guidance to help it direct the installation of "flood safe" primary and secondary side makeup systems. The guidance currently requires a large number of tasks in potentially challenging environmental conditions to prepare the plant for extreme flooding conditions. Review of the guidelines has resulted in the identification of several areas that could be improved to reduce the time required to implement the procedures and to improve the reliability of the process.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-217 License Renewal Application While the details of the enhancements have not yet been developed, the following high level improvements have been established as desirable for inclusion:
* Fully proceduralize guidelines:  Upgrade the guidelines so that they provide step by step instructions on all aspects of the implementation process. For example, the guidelines currently direct connections to power and air sources without specifying the steps required to complete the connection. The details for these types of tasks must be provided,
* Permanently mount the power cables between the generator and pump staging areas,
* Permanently mount the emergency seal injection pump with a suction source from the fuel transfer tubes and use it in place of the submersible injection pumps to take advantage of its capability of injecting at normal operating pressure (rather than the 1200 psig available from the submersible pumps). The pump must be positioned at a flood-safe height,
* Permanently mount injection lines required for primary and secondary side makeup (may not be practical for the secondary side pump that takes suction from flood water in the turbine
 
building),
* Consider an alternate secondary side suction source given that flood waters may recede well before an alternate secondary side makeup source will become available when AC power is re-established to the site,
* Ensure the power cables have all required connectors attached or stored in the staging areas,
* Pre-manufacture any required air supply or fuel oil connectors and store them in the staging areas,
* Store the portable generator on the turbine deck,
* Install a normally empty fuel oil tank for EG-Y-6 on the turbine deck that can be filled when it is required using power from EG-Y-6, if necessary. Based on the IPEEE evaluation, one of the larger contributors to the severe flooding mitigation strategy is the reliability of EG-Y-6. For the 48 hour mission time evaluated in the relevant external flooding scenarios, the failure probability for EG-Y-6 is nearly 1.5E-01. This estimate is Environmental Report Appendix E  SAMA ANALYSIS Page E-218 Three Mile Island Nuclear Station Unit 1 License Renewal Application based on the use of the failure probabilities established for the large 4kV EDGs used to power the emergency buses at TMI. While the use of the EDG failure data for EG-Y-6 is believed to be conservative, component specific failure data for EG-Y-6 is not available. As a result, the design of this SAMA requires the purchase of an alternate, diverse portable generator to serve as a backup AC power source. E.6.32.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk
 
associa t ed with the non-external flooding events.
As desc r ibed in Section E.4.6.3 , the external events risk, excluding external flooding, is considered to be equal to the internal events risk. Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise
 
the total averted cost-risk for a SAMA. In this case, the changes to the extreme flooding mitigation strategy are not expected to impact internal events or non-flooding external events risk. This is because the primary injection alignment cannot be performed before RCP seal heatup/damage in SBO events or other scenarios that lead to loss of seal cooling. For a majority of the external flooding cases, the severe flooding primary injection strategy could be aligned before the loss of on-site AC power such that seal cooling would never be lost. For the internal events model, there is no adequate warning that would allow such an early alignment and the only result from using the severe flooding primary injection method would likely be thermal shock to the RCP seals. Based on the discussion above, this SAMA does not reduce internal events risk, as summarized below: SAMA 32 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.37E-05 32.61 $112,259 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-219 License Renewal Application SAMA 32 - Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 The non-external flooding external events contribution is typically calculated using the 2.0 multiplier on the internal events results, but in this case, the averted cost-risk is $0, so the non-external flooding external events contribution is also $0: SAMA 32 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,271,711 $0 2.0 $0 E.6.32.2 EXTERNAL FLOODING EVALUATION The severe flooding guidelines were originally credited in the IPEEE for both floods above 310' msl as well as for floods between 305' and 310' msl. Due to a more limited preparation time for the 305' to 310' msl floods, the failure probability was assumed to be 0.5 rather than the 0.255 used for the 310' msl floods. For floods below 305' msl, no credit was taken for the severe flooding guidelines as the submersible pumps used for secondary side makeup require flood water in the turbine building for a suction source. While this SAMA includes provisions for an alternate secondary side pump suction source, t he expected alignment time of approximately 2 hours would likely preclude it from being an effective means of preventing core damage in a flood induced loop. In these cases, the alignment of the severe flood equipment would not begin in time to establish cooling before core melt.
Environmental Report Appendix E  SAMA ANALYSIS Page E-220 Three Mile Island Nuclear Station Unit 1 License Renewal Application For the purposes of this analysis, implementation of this SAMA is assumed to reduce the HEP for alignment of the external flooding measures from 1.1E-01 to 1.0E-02. In addition, the availability of the diverse, alternate portable AC generator is considered to reduce the failure probability of the flood-safe AC power source from 1.43E-01 to 2.04E-02 (1.43E-01
* 1.43E-01
= 2.04E-2, which assumes completely independent generators). This results in a total failure
 
probability of 3.04E-02 (1.0E-02 + 2.04E-02 = 3.04E-02) for the severe flooding mitigation
 
strategy. Because the severe flooding guidelines were credited differently in each of the flood ranges, three separate strategies are required to obtain the revised core damage frequencies for the flooding scenarios:
* Floods >310' msl:  The CDF for this scenario was calculated in the IPEEE as the product of the flood frequency and the failure probability for the alignment of the severe flooding mitigation strategy. As a result, the revised frequency can be obtained by multiplying the base CDF by the ratio of SAMA based severe flood mitigation failure probability to the baseline severe flood mitigation failure probability (3.04E-02 / 2.55E-01 = 1.19E-01).
* Floods between 305' and 310' msl:  In the IPEEE, a multiplier of 0.5 was applied to each of the sequences in the flooding event tree to represent the potential to avert the flood using the severe flooding guidelines. The increase in the failure probability over the >310' msl case was made to account for the decreased time available in the 305' and 310' msl cases.
For this evaluation, it is assumed that the failure to implement SAMA 32 is dominated by operator dependence. Non-negligible dependence exists between the actions to install the flood gates and to implement SAMA 32; however, the dependence is cognitive. As execution failure would be the majority contributor to the flood HEPs and because the execution and cognitive contributors are not separated for the flood actions, it could be overly conservative to use the dependence factors based on the available HEPs, especially given that the appropriate dependence level would likely be "high". To simulate the results of a true dependence assessment where the cognitive and execution components of the HEP are explicitly provided, a moderate dependence factor is used rather than a high dependence factor. As a result, the base event tree failure probabilities are multiplied by 0.14 (which is obtained from equation 10-16 of NUREG/CR-1278 (NRC 1983)) rather than 0.5 to get the new frequencies. In order to obtain the post-SAMA frequencies for these sequences, the flood frequencies reported in the IPEEE are multiplied by 0.28 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-221 License Renewal Application (0.14/0.5=0.28) to account for the original 0.5 failure probability assigned to the contemporary severe flooding guidelines. Sequence "F" represents failure of the early flood warning and precluded the use of the flood panels in the IPEEE; however, credit was taken for the severe flood guidelines in the same manner as for the other sequences. For consistency with the IPEEE, the same credit taken for SAMA 32 in the other 305' to 310' msl sequences is also taken for Sequence "F". Given that Sequence "F" is a minimal contributor, this assumption has no meaningful impact on the results.
* Floods below 305' msl:  While there are provisions to use a non-floodwater suction source for SAMA 32, the 2 hour alignment time may make it ineffective to prevent CD after flood induced LOOP. No credit is taken for SAMA 32 for these floods. The results of this process are summarized below:
SAMA 32 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 1.26E-05 28.5 $87,324 Percent Change
-84.4% -83.6% -83.9% A further breakdown of this information is provided below according to release category. SAMA 32 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 7.58E-06 1.76E-06 1.86E-08 2.53E-07 1.71E-06 1.02E-06 2.42E-08 2.50E-07 1.26E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 15.80 3.68 0.05 0.53 5.00 3.00 0.07 0.37 28.50 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $48,308 $11,242 $167 $1,615 $15,355 $9,200 $218 $1,219 $87,324 The external flooding based averted cost-risk for this SAMA is shown below:
Environmental Report Appendix E  SAMA ANALYSIS Page E-222 Three Mile Island Nuclear Station Unit 1 License Renewal Application SAMA 32 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding  Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $2,491,451 $13,052,022 E.6.32.3 COST OF IMPLEMENTATION The cost of implementation is estimated to be $1,700,000 (Exelon 2007c). E.6.32.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 32 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $0 $13,052,022 $13,052,022 $1,700,000 $11,352,022 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net value is positive. E.6.33 SAMA NUMBER 33:  INCREASE THE FLOOD PROTECTION HEIGHT The current configuration protects to the design basis limit of 310 feet msl and levels any higher result in topping of the existing flood doors and flooding of sensitive areas. Raising the height of the flood doors (completely sealing the doors, raising required air intakes/exhaust ducts, as required) would prevent water incursion and allow for continued operation of the normal safety equipment. The goal of this SAMA is to increase the flood protection height to a point where the extreme flooding CDF would be comparable to the internal events CDF of 2.37E-05/yr. In this case, the goal is assumed to be a CDF of 1.0E-05/yr and the assumption is made that when the flood waters exceed the flood protection height, core damage occurs (no credit taken for existing extreme flooding guidance). The exceedance frequency of 1E-05/yr corresponds to a level of 320.3' msl (stillwater level). Protecting the plant against these floods would require modifications to match the stillwater of 320.3' msl plus the wave setup height, which was Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-223 License Renewal Application determined to be up to 4' (GPU 1990). The total flood protection height required is, therefore, 324.3' msl, which is rounded up to 324.5' msl. Based on a review of plant documentation, the following changes would be required to protect the plant up to 324.5' msl:
* EDG Building, GATE D1:  Raise the flood gate to completely seal the door.
* EDG Building, GATE D2:  Sealed by security changes, no additional changes are required.
* EDG Building, GATE D3:  Raise the flood gate to completely seal the door.
* EDG Building, GATE D4:  The gates must be extended from 311'-0" to 324.5'.
* EDG Building, Air Vent Valve:  A 2-1/2" diameter penetration is located at elevation 311'-2" in the north wall and one in the west wall at elevation 312'-4" for fuel oil day tank air vent valves. Both penetrations must be waterproofed and the outlets must be extended to 324.5'.
* Air Intake Structure, Access Door:  The bottom of the door is at 312'-0" and has no flood protection. The door should be completely sealed.
* Air Intake Structure, Air Vents:  The bottoms of the intake louvers are a 312'. These must be completely sealed.
* Intermediate Building, Gate C-1:  The tops of the existing gates are at 311'-6" and leave about 3 feet open to the top of the doorway. The door should be completely sealed and fitted with an entry hatch.
* Control Building, Gate B-1:  The tops of the existing gates are at 311' and leave an additional 10 feet open to the top of the doorway. The doors should be sealed and fitted with an entry hatch. Covering the entire doorway may not be required, but the conservative modification would be to provide complete protection.
* Control Building, Gate B-2:  The tops of the existing gates are at 311' and leave an additional 2 feet open to the top of the doorway. The door should be completely sealed and fitted with an entry hatch.
Environmental Report Appendix E  SAMA ANALYSIS Page E-224 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* Intake Screen Pumphouse, Gate E-1:  The tops of the existing gates are at 311' and leave an additional 4 feet open to the top of the doorway. The door should be completely sealed and fitted with an entry hatch.
* Intake Screen Pumphouse, Gate E-4:  The tops of the existing gates are at 311' and leave an additional 6 feet open to the top of the doorway. The door should be completely sealed
 
and fitted with an entry hatch.
* Intake Screen Pumphouse, Gate E-2:  The tops of the existing gates are at 311' and leave an additional 4 feet open to the top of the doorway. The door should be completely sealed
 
and fitted with an entry hatch.
* Intake Screen Pumphouse, Gate E-3:  The tops of the existing gates are at 311' and leave an additional 4 feet open to the top of the doorway. The door should be completely sealed and fitted with an entry hatch. In addition, two penetrations exist at 311'-4" and 312'-8" and an exhaust penetration exists in the west wall. These penetrations must be sealed and communication with the atmosphere must be provided at a level of at least 324.5'. In addition, there is concern that the cable vaults holding the cables that connect the EDGs to the emergency buses are not waterproof. These cable vaults must be waterproofed so that the EDG output cables do not short out in the event that the cables have lost integrity. E.6.33.1 INTERNAL EVENTS AND NON-EXTERNAL FLOODING EVALUATION This subsection describes the calculation of the component of this SAMA's averted cost-risk associa t ed with the non-external flooding events.
As desc r ibed in Section E.4.6.3 , the external events risk, excluding external flooding, is considered to be equal to the internal events risk.
Quantitatively, this is accounted for by multiplying the internal events averted cost-risk by a factor of 2.0. This process is described below and is one of the two components that comprise
 
the total averted cost-risk for a SAMA. In this case, the changes to the extreme flooding protection height will not impact internal events or non-flooding external events risk and this SAMA will not reduce the CDF, dose risk, or OECR, as summarized below:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-225 License Renewal Application SAMA 33 - Internal Events Results CDF (/yr) Dose-Risk OECR Base Results 2.37E-05 32.61 $112,259 SAMA Results 2.37E-05 32.61 $112,259 Percent Change 0.0% 0.0% 0.0% A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06. SAMA 33 Internal Events Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)SAMA 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskSAMA 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRSAMA $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)SAMA 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskSAMA 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRSAMA $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 The non-external flooding external events contribution is typically calculated using the 2.0 multiplier on the internal events results, but in this case, the averted cost-risk is $0, so the non-external flooding external events contribution is also $0: SAMA 33 - Non-External Flooding Averted Cost-Risk Base Case Internal Events Cost-Risk SAMA Case Internal Events Cost-Risk Internal Events Averted Cost-Risk Non-Flood External Events Multiplier Total Non-FloodAverted Cost-Risk $3,271,711 $3,271,711 $0 2.0 $0
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-226 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.6.33.2 EXTERNAL FLOODING EVALUATION This SAMA only has the potential of reducing the risk of the extreme floods, those which result in flood waters exceeding 310' msl. The lesser fl oods are already protected by dikes or flood gates and for those cases where flood gate installation fails, this SAMA would also fail. For the purposes of this analysis, implementation of this SAMA is assumed to eliminate extreme flooding risk if installed properly. The same failure probability used in the IPEEE for installing the flood doors for the 305' to 310' msl floods is used for the floods over 310' (HSL1 at 5.62E-02). In this case, no credit is taken for the implementing the existing severe flooding guidelines in the event that SAMA 33 is implemented and fails. SAMA 33 impacts neither the 305' to 310' msl floods nor the floods below 305' msl. The results of this assumption are summarized below:  SAMA 33 - External Flooding Results CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 SAMA Results 3.14E-5 73.59 $225,428 Percent Change
-61.3% -58.5% -58.4% A further breakdown of this information is provided below according to release category. SAMA 33 - External Flooding Contributions by Release Category Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B305' to 310' sequence C305' to 310' sequence D305' to 310' sequence E305' to 310' sequence F
<305' (uses LOOP RC distribution)Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 SAMA Frequency 1.40E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 3.14E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 SAMA Dose-Risk 29.18 13.13 0.19 1.89 17.87 10.71 0.25 0.37 73.59 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 SAMA OECR $89,220 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $225,428 The external flooding based averted cost-risk for this SAMA is shown below: SAMA 33 - External Flooding Averted Cost-Risk Base Case External Flooding Cost-Risk SAMA Case External Flooding Cost-Risk External Flooding Averted Cost-Risk $15,543,473 $6,401,188 $9,142,285
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-227 License Renewal Application E.6.33.3 COST OF IMPLEMENTATION The cost of this enhancement was estimated to be $2,700,000 by the TMI staff (Exelon 2007c). E.6.33.4 NET VALUE The net value for this SAMA is the difference between the total averted cost-risk, which is the sum of the external flooding and non-external flooding based averted cost-risks, and the cost of implementation. The following table summarizes these results: SAMA 33 - Net Value Non-External Flooding Based Averted  Cost-Risk External Flooding Based Averted Cost-Risk Total Averted Cost-Risk Cost of Implementation Net Value $0 $9,142,285 $9,142,285 $2,700,000 $6,442,285 Given that the cost of implementation is less than the averted cost-risk for this SAMA, the net
 
value is positive.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-228 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.7 UNCERTAINTY ANALYSIS Sensitivity cases were run for the following conditions to assess their impact on the overall SAMA evaluation:
Use the 95 th percentile PRA results in place of the mean PRA results. Assume no baseline BWST Refill capability
* Use alternate MACCS2 input variables for selected cases.
* Assume no credit for extreme external flooding guidance E.7.1 95 TH PERCENTILE PRA 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
 
95 th percentile results to examine the impact of uncertainty in the PRA model. For TMI-1, the UNCERT32 software code was used to perform the Level 1 internal events model uncertainty analysis. The results of the calculation are provided below:
Parameter Value Mean 4.10E-05 5 percent 8.98E-06 50 percent 1.81E-05 95 percent 6.51E-05 Standard Deviation 9.36E-04  The PRA uncertainty calculation identifies the 95 th percentile CDF as 6.51E-05 per year. This is a factor of 2.75 greater than the CDF point estimate produced by the TMI-1 PRA.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-229 License Renewal Application E.7.1.1 PHASE I IMPACT For Phase I screening, use of the 95 th percentile PRA results will increase the MACR and for some sites, it may prevent the screening of some of the higher cost modifications. In the event that a SAMA is retained based on use of the 95 th percentile MACR, it would be unlikely to impact the SAMA conclusions. This is due to the fact that the benefit gleaned from the implementation of those SAMAs must be extremely large in order to be cost beneficial. For TMI-1, no SAMAs were screened in Phase I, so use of the 95 th percentile PRA results does not impact the Phase I analysis. However, the 95 th percentile PRA results MACR is calculated here for completeness. As discussed above, the 95 th PRA results are approximately a factor of 2.75 greater than the point estimate CDF. The uncertainty analyses that are available for the Level 1 models are not available for Level 2 and 3 PRA models. In order to simulate the use of the 95 th percentile results for the Level 2 and 3 models, the same scaling factor calculated for the Level 1 results was assumed to apply to the Level 2 and 3 models. Because the MACR calculations scale linearly with the CDF, dose-risk, and offsite economic cost-risk, the 95 th percentile MACR can be calculated by multiplying the base case MACR by 2.75. This results in a 95 th percentile MACR of $60,739,250. E.7.1.2 PHASE II IMPACT As mentioned above, the 95 th percentile PRA results are not available for the Level 2 and 3 models. In order to estimate the impact of using the 95 th percentile PRA results in the Phase II SAMA analysis, the same process used to calculate the revised MACR was applied to each of the Phase II SAMAs (the averted cost-risk for each SAMA was increased by a factor of 2.75 over the base case). The following table provides a summary of the impact of using the 95 th percentile PRA results in the detailed cost-benefit calculations that have been performed.
Environmental Report Appendix E  SAMA ANALYSIS Page E-230 Three Mile Island Nuclear Station Unit 1 License Renewal Application Results Summary for the 95 th Percentile PRA Results SAMA ID Cost of Implement-ation Averted Cost- Risk(Base) Net Value (Base) Averted Cost- Risk(95th Percentile) Net Value (95th Percentile) Change in Cost Effective-ness? SAMA 1 $3,125,000 $986,145 -$2,138,855 $2,711,899 -$413,101 No SAMA 2 $7,300,000 $4,297,001 -$3,002,999 $11,816,753 $4,516,753 Yes SAMA 3 $2,450,000 $594,926 -$1,855,074 $1,636,047 -$813,954 No SAMA 5 $3,150,000 $230,163 -$2,919,837 $632,948 -$2,517,052 No SAMA 6 $2,750,000 $398,924 -$2,351,076 $1,097,041 -$1,652,959 No SAMA 7 $1,000,000 $449,254 -$550,746 $1,235,449 $235,449 Yes SAMA 8 $145,000 $1,234,676 $1,089,676 $3,395,359 $3,250,359 No SAMA 10 $3,800,000 $982,048 -$2,817,952 $2,700,632 -$1,099,368 No SAMA 11 $4250,000 $16,088,692 $11,838,692 $44,243,903 $39,993,903 No SAMA 12 $50,000 $198,438 $148,438 $545,705 $495,705 No SAMA 13 $950,000 $305,294 -$644,706 $839,559 -$110,442 No SAMA 14 $3,150,000 $603,886 -$2,546,114 $1,660,687 -$1,489,314 No SAMA 15 $450,000 $199,098 -$250,902 $547,520 $97,520 Yes SAMA 16 $1,100,000 $1,592,631 $492,631 $4,379,735 $3,279,735 No SAMA 17 $950,000 $51,988 -$898,012 $142,967 -$807,033 No SAMA 18 $100,000 $33,260 -$66,740 $91,465
-$8,535 No SAMA 19 $760,000 $3,127,876 $2,367,876 $8,601,659 $7,841,659 No SAMA 20 $3,030,000 $173,974 -$2,856,026 $478,429 -$2,551,572 No SAMA 21 $1,200,000 $1,181,137 -$18,863 $3,248,127 $2,048,127 Yes SAMA 22 $5,000,000 $1,253,768 -$3,746,232 $3,447,862 -$1,552,138 No SAMA 23 $50,000 $30,629 -$19,371 $84,230
$34,230 Yes SAMA 24 $8,400,000 $4,416,201 -$3,983,799 $12,144,553 $3,744,553 Yes SAMA 25 $6,000,000 1,466,139 -$4,533,861 $4,031,882 -$1,968,118 No SAMA 26 $900,000 $369,663 -$530,337 $1,016,573 $116,573 Yes SAMA 27 $575,000 $1,306,819 $731,819 $3,593,752 $3,018,752 No SAMA 28 $575,000 $47,891 -$527,109 $131,701 -$443,299 No SAMA 29 $800,000 $25,716 -$774,284 $70,719 -$729,281 No SAMA 30 $150,000 $25,716 -$124,284 $70,719 -$79,281 No SAMA 32 $1,700,000 $13,052,022 $11,352,022 $35,893,061 $34,193,061 No SAMA 33 $2,700,000 $9,142,285 $6,442,285 $25,141,284 $22,441,284 No Of the SAMAs classified as "not cost beneficial" in the baseline Phase II analysis, seven SAMAs (2, 7, 15, 21, 23, 24, and 26) were found to be cost beneficial when the 95 th percentile PRA results were applied. The use of the 95 th percentile PRA results is not considered to provide the most realistic assessment of the cost effectiveness of a SAMA; however, these seven additional SAMAs could be considered for implementation to address the uncertainties inherent in the
 
SAMA analysis.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-231 License Renewal Application E.7.2 BWST REFILL CAPABILITY A recent inspection at TMI questioned the viability of preventing core damage in SGTR scenarios by refilling the BWST. Specifically, it is not certain whether the BWST can be refilled at a rate that will completely make up for the inventory being lost through the tube rupture.
Analysis has shown that in certain scenarios (e.g., no RCS cooldown and depressurization), the current BWST refill capability will only delay core damage, but not prevent it. Because SGTR events are large contributors to the TMI-1 dose-risk and OECR, changes to the assumptions related to BWST refill capabilities can have a significant impact on the accident consequence analysis given that successful BWST refill is assumed to avert core damage. While the results of the BWST refill analysis have not yet been finalized, this sensitivity has been developed to determine how the SAMA 10 evaluation (automated BWST refill) could be impacted by the assumption that manual BWST refill is not capable of preventing core damage for SGTR events at TMI-1. Currently, the PRA model assumes that manual refill of the BWST will support continuous makeup to the primary system, thus preventing core damage in SGTR scenarios. The importance list review showed that further improving the reliability of this function would have a meaningful impact on both the Level 1 and Level 2 results. The cost benefit results provided for SAMA 10 in Section E.6.10 are pred i cated on the assumption that the cu rrent BWST refill capability prevents core damage; however, if the current capability only delayed core damage to allow other recovery actions rather than pr event core damage, the impact of implementing SAMA 10 would be greater than what is shown in the baseline assessment. The averted cost-risk for the SAMA would be estimated using the difference in the MACR for the plant configuration in which BWST refill always fails and the MACR for the plant configuration in which BWST is fully automated. This is a bounding assessment since assuming no refill capability is conservative. However, detailed modeling of partial success via manual BWST refill would be very complicated and may only reduce the averted cost-risk by a small amount. Because SAMA 10 already evaluated the plant configuration in which BWST refill is fully automated, the information required to obtain the MACR for that plant configuration is already available and is the sum of the "SAMA case external flooding cost risk" and 2 times the "SAMA case internal events cost-risk" (multiplier of 2 required to account for the non-external flooding external ev e nts contribu t ion). As documented in Section E.6.10 , the "SAMA case ex t ernal Environmental Report Appendix E  SAMA ANALYSIS Page E-232 Three Mile Island Nuclear Station Unit 1 License Renewal Application flooding cost risk" is $15,543,473 and the "SAMA case internal events cost-risk" is $2,763,004. The total MACR would therefore be $21,069,481 ($15,543,473 + 2 * $2,763,004). In order to obtain the revised baseline MACR in which BWST refill always fails, the basic event representing the independent operator action for BWST refill is set to 1.0. Because failure of the BWST refill action is a physical limitation, the JHEPs are eliminated from the results. The changes made to the cutset file to obtain the "revised baseline" results are summarized in the table below: BWST Refill Sensitivity Model Changes Gate and / or Basic Event ID and Description Description of Change BWST-HRE27-HTKOA:  FAILURE TO REFILL BWST (SPLIT FRAC REV) The basic event probability was changed from 2.65E-02 to 1.0. JHAHCD4RE27HEPOA: AVHCD4_FF--HCDOA AND BWST-HRE27-HTKOA (JHEP addressing BWST refill and cooldown via secondary side) The basic event probability was changed from 9.17E-05 to 0.0.
JHHRE27HL1AHEPOA: BWST-HRE27-HTKOA AND DLHHL1A----HVHOA (JHEP addressing BWST refill and opening drop line for DHR cooling) The basic event probability was changed from 2.00E-04 to 0.0. JHHEF2HRE27HEPOA: AVHEF2_FF--HCDOA AND BWST-HRE27-HTKOA (JHEP addressing BWST refill and manually initiating cooldown using the OTSG) The basic event probability was changed from 1.3E-03 to 0.0. JHHIGHREHHLHEPOA: IGHIG1_HER-HSGOA, BWST-HRE27-HTKOA, and DLHHL1A----HVHOA (JHEP addressing BWST refill, failure to isolate a SGTR, and opening drop line for DHR cooling) The basic event probability was changed from 5.0E-07 to 0.0. (Event was not in cutsets) The results of these changes are summarized in the tables below: BWST Refill Sensitivity Results CDF (/YR) DOSE-RISK OECR Sensitivity Results 2.75E-05 54.04 $216,329 A further breakdown of this information is provided below according to release category. Note that the results for the following RCs are not provided given that the frequencies are always zero: RC2-01, RC2-03, RC3-05, RC3-06, RC4-05, RC4-06, RC4-07, RC4-08, RC6-01, RC6-02, AND RC6-06.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-233 License Renewal Application BWST Refill Sensitivity Results By Release Category Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)Sens 2.33E-06 3.46E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskSens 13.33 19.79 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRSens $64,774 $96,188 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)Sens 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.33E-051.69E-08 2.36E-06 1.91E-082.75E-05 Dose-RiskSens 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.54 0.00 0.63 0.01 54.04 OECRSens $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,476 $4 $618 $5 $216,329 These resul t s are converted into a c o st-risk using the methods documented in Secti o n E.4: BWST Refill Sensitivity Non-External Floodi ng Cost-Risk Sensitivity Case  Internal Events  Cost-Risk Non-External Flooding External Events Multiplier Total Non-Flood Cost-Risk $5,578,084 2.0 $11,156,168 Assuming that the external flooding MACR is constant at $15,543,473, the total MACR for the case without BWST refill capability would be $26,699,641 ($15,543,473 + $11,156,168 = $26,699,641). It should be noted that the use of the multiplier of 2 to account for external events contributions for this case may be inappropriate because SGRT events are not
 
considered in the external events scenarios. Finally, the averted cost risk and net value for SAMA 10 assuming an initial configuration in which BWST refill is not credited can be recalculated: BWST Sensitivity SAMA 10 Net Value No BWST Refill Case MACR Fully Automated BWST Refill MACRAverted Cost-Risk Cost of Implementation Net Value $26,699,641 $21,069,481 $5,630,160 $3,800,000 $1,830,160 Given that the net value is positive for this case, the implication is that if the actual TMI-1 conditions are best represented by no credit for BWST refill (a conservative assumption), SAMA 10 is a cost effective change.
Environmental Report Appendix E  SAMA ANALYSIS Page E-234 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.7.3 MACCS2 INPUT VARIATIONS The MACCS2 model was developed using the best in formation available for the Three Mile Island 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, a sensitivity analysis was performed on a group of parameters that has previously been shown to impact the Level 3 results. These parameters (and the associated sensitivity case identifiers)
 
include:
* Meteorological data (TMI1999; TMI2000)
* Population estimates(TMI30INC; TMISIT00)
* Evacuation effectiveness (TMISLOW)
* Radionuclide release characteristics (TMIATM1; TMIATM2)
* Recovery, decontamination, and resettlement factors (Intermediate Phase) (TMICHR1, TMICHR2) The risk metrics produced by MACCS2 that are evaluated in the sensitivity analyses are the 50 mile population dose and the 50 mile offsite economic cost. The following subsections discuss the changes in these results for each of the sensitivity cases that are shown below. The final
 
subsection, E.7.3.6 , correlates the w orst case c h anges identified in the sensitivity runs to a change in the site's averted cost-risk and discusses the implications of the sensitivity analysis on the SAMA analysis. CASE DESCRIPTION POP.
DOSE RISK  BASE (%) COST RISK BASE (%) TMI1998 Base Case (Year 1998 MET data)
-- -- TMI1999 Year 1999 MET data -10.5% -9.29% TMI2000 Year 2000 MET data -4.73% -5.15% TMI30INC Year 2034 population values increased uniformly 30% over base case.
28.3% 29.5% TMISit00 Year 2000 population based (Base Case is Year 2034) -28.9% -29.6% TMISlow Evacuation speed decrea sed 50% to 0.59 mph, 0.26 m/sec (Base Case is 1.18 mph).
15.3% 0% TMIATM1 Release height set to ground level -4.58% -5.22%
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-235 License Renewal Application CASE DESCRIPTION POP. DOSE RISK  BASE (%) COST RISK BASE (%) TMIATM2 Plume thermal heat content set to ambient (i.e., buoyant plume rise not modeled) 1.65% 1.09% TMICHR1 Long Term Phase starts immediately after the Early Phase is over (No Intermediate Phase; Base Case is 6 month Intermediate Phase) 16.8% -36.9% TMICHR2 1 Year Intermediate Phase following the Early Phase (Base Case is 6 month Intermediate Phase)
-8.84% 34.0%
E.7.3.1  METEOROLOGICAL SENSITIVITY In addition to the base case meteorological data (year 1998), data is also analyzed for the years 1999 and 2000. Analysis of these alternate data sets yielded population dose-risks and offsite economic cost-risks that are lower than the 1998 data by at least 4.7 percent and by as much as 10.5 percent. As no particular criteria have been defined by the industry related to determining which meteorological data set should be used as a base case for a site, the year 1998 data is conservatively chosen for Three Mile Island given that it yielded the largest results. E.7.3.2 POPULATION SENSITIVITY The population sensitivity cases (TMI30INC, TMISIT00) demonstrate a significant dependence on population estimates. This is expected given that the population dose and offsite economic costs are primarily driven by the regional population. In case TMI30INC, the baseline 2034 population is uniformly increased by 30 percent in all sectors of the 50-mile radius. This change increased the estimated population dose-risk and offsite economic cost by over 28 percent each. A second population based sensitivity (TMISIT00) is performed to determine the impact of using year 2000 census data rather than projecting to the end of the license renewal period (Year 2034). The baseline SAMA case is based on a population projection to year 2034 based on the population growth trends shown between the years 1990 and 2000. When year 2000 data is utilized, the overall dose-risk and OECR decrease, as expected. Specifically, the dose-risk and the OECR decreased by about 29 percent each.
Environmental Report Appendix E  SAMA ANALYSIS Page E-236 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.7.3.3 EVACUATION SENSITIVITY The evacuation sensitivity case (TMISLOW) demonstrates population dose-risk impacts associated with evacuation assumptions. While evacuation assumptions do impact the population dose-risk estimates, they do not impact MACCS2 offsite economic cost-risk estimates because MACCS2 calculated cost-risks are based on land contamination levels which remain unaffected by evacuation assumptions and the number of people evacuating. For Three Mile Island, a slow evacuation assumption is used in the base case (1.18 mph). An additional 50 percent decrease in the evacuation speed to 0.59 mph increased the dose-risk by approximately 15 percent. E.7.3.4 RADIOACTIVE RELEASE SENSITIVITY The sensitivity cases TMIATM1 and TMIATM2 quantify the impact of the assumptions related to the height of the release and thermal energy of the plume, respectively. TMIATM1 assumes that the release occurs at ground level rather than at an elevation that could correspond to a release through the stack or a break high in the reactor building. The lower release height shows a decrease in dose-risk and OECR of approximately 5 percent. Reducing the thermal plume heat content to ambient conditions has a minimal impact. TMIATM2 shows an increase in the dose-risk and the OECR of about 1 percent. E.7.3.5 INTERMEDIATE PHA SE DURATION SENSITIVITY The Intermediate Phase, as modeled by MACCS2, is the time period beginning after the early phase (one week emergency phase) and extends to the time when recovery actions such as decontamination and resettlement are started (long term phase). MACCS2 allows the habitation of land during the intermediate phase unless the projected dose criterion is exceeded.
If the projected dose criterion is exceeded during the intermediate phase, the individual is relocated. MACCS2 allows an intermediate phase ranging from no intermediate phase to one (1) year. The Intermediate Phase related sensitivity cases (TMICHR1 and TMICHR2) show significant dependence in relation to economic impact, and are therefore discussed further:
* The No Intermediate Phase case (TMICHR1) is developed based on the NUREG-1150 modeling approach. However, the 37 percent reduction in economic cost estimates based on the approach are judged too optimistic in that the land decontamination efforts are modeled as starting one week after the accident (i.e., directly after the early phase ends)
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-237 License Renewal Application such that a significant portion of population relocation costs are omitted. For example, the costs associated with temporary housing while decontamination strategies are developed and decontamination teams are contracted are not accounted for without an intermediate phase. It is believed that NUREG-1150 studies omitted the intermediate phase because the MACCS2 intermediate phase coding was not validated at that time. A competing factor is that the population dose increases because people are allowed to re-occupy the land sooner (17 percent increase over the base case).
* The 1 Year Intermediate Phase case (TMICHR2) is developed based on the maximum length of time allowed by MACCS2 for the intermediate phase. A long intermediate phase can be unrealistic in that re-occupation of the contaminated land is not performed during this phase even if contamination levels decrease (by natural radioactive decay) to levels which would allow it (i.e., resettlement is evaluated as part of the long term phase, not the intermediate phase). Therefore, population relocation costs may be over estimated using a long (i.e., one year) intermediate phase. An Intermediate Phase of one year shows a 34 percent increase in the OECR estimates compared with the six month (base case)
Intermediate phase. However, the population dose decreased by 9 percent with a longer Intermediate Phase due to later resettlement on decontaminated land. The six month intermediate phase (base case) is judged to be a best estimate approach in that it provides a reasonable time for both decontamination efforts and resettlement to begin. The sensitivity cases demonstrate that this six month modeling approach is mid-range of the modeling choices available and is used as the base case. E.7.3.6 IMPACT ON SAMA ANALYSIS Several different Level 3 input parameters are examined as part of the Three Mile Island MACCS2 sensitivity analysis. The primary reason for performing these sensitivity runs is to identify any reasonable changes that could be made to the Level 3 input parameters that would
 
impact the conclusio n s o f the SAMA analysis.
W hile the table in Section E.7.3 summarizes the changes to the dose-risk and OECR estimates for each sensitivity case, it is prudent to consider if any of these changes would result in the retention of the SAMAs that were screened using the baseline results. Of all the MACCS2 sensitivity cases, the largest increase in the dose-risk is 28 percent in case TMI30INC while the largest increase in OECR is 34 percent in case TMICHR2. While these are Environmental Report Appendix E  SAMA ANALYSIS Page E-238 Three Mile Island Nuclear Station Unit 1 License Renewal Application separate cases, the Three Mile Island MACR is recalculated using these results to determine the impact of using the worst case for each parameter simultaneously. The resulting MACR is $28,048,743 (a factor of 1.27 increase over the base case), which is less than the $60,739,250 calculated in Section E.7.1 for the 9 5 th percentile PRA results. The 9 5 th p ercentile PRA results sensitivity is considered to bound this case and no SAMAs would be retained based on this
 
sensitivity t h at were not already identified in Sec t ion E.7.1. E.7.4 EXTREME FLOODING MITIGATION The extreme flooding scenario (floods over 310' msl) accounts for 53% of the TMI-1 MACR. While this single sequence is highly important to site risk, the calculation of its CDF is simplified, using only the frequency of a flood exceeding 310' msl (stillwater level) and the failure probability of the severe flood mitigation strategy. In addition, the flooding sequences between 305' and 310' msl contribute a CDF of 1.71E-05/yr and are also based on simplified risk estimates. Typically, simplified estimates such as these include a conservative bias to prevent under predicting negative events; however, due to the large uncertainty in external flood scenarios, it is still possible that the quantification results underestimate the flooding risk. This sensitivity is intended to examine how an optimistic assessment of the flooding risk could impact the SAMA analysis. This sensitivity could be accomplished by modifying the flooding frequency for each of the flood ranges by a set factor, but in this case, the source of uncertainty was assumed to be in the likelihood of successfully implementing the extreme flooding mitigation strategy, which is credited for both floods over 310' msl and those between 305' and 310' msl.
SAMA 32 investigates the cost benefit of improving the extreme flooding mitigation strategy, but this sensitivity will provide some insight on how the existing assumptions related to the response capability impact the other SAMA evaluations. E.7.4.1 PHASE I IMPACT In this sensitivity, no credit is taken for the use of the current TMI severe flood guidance. This pessimistic assumption changes the extreme flooding CDF from 6.37E-05/yr to 2.50E-04/yr. In addition, the CDFs for all of the sequences in the 305' msl to 310' msl range are increased by a factor of two, which mathematically eliminates the credit taken for the flood guidelines for those sequences. The following table summarizes the changes to the dose-risk and OECR corresponding to these changes in CDF:
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-239 License Renewal Application Flooding Sensitivity: No Credit for Severe Flooding Guidance CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Base Ext. Flooding Results 8.11E-05 177.16 $542,159 Sensitivity Results 2.84E-04 609.47 $1,864,412 Percent Change +250.2% +244.0% +243.9% A further breakdown of this information is provided below according to release category. Flooding Sensitivity: Contributions by Sequence Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood FrequencyBase Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Sensitivity Frequency 2.50E-04 1.26E-05 1.33E-07 1.81E-06 1.22E-05 7.31E-06 1.73E-07 2.50E-07 2.84E-04 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Sensitivity Dose-Risk 521.00 26.26 0.39 3.77 35.75 21.42 0.51 0.37 609.47 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159 Sensitivity OECR $1,593,214 $80,298 $1,196 $11,535 $109,678 $65,717 $1,555 $1,219 $1,864,412 The corresponding external flooding component of the averted cost-risk is shown below: Flooding Sensitivity: Revised External Flooding MACR Base Case External Flooding MACR Sensitivity Case External Flooding MACR Difference (Sensitivity MACR - Base MACR)  $15,543,473 $53,604,345 $38,060,872 As can be seen, assuming no credit for TMI's current extreme flood mitigation capabilities results in a large increase in the external flooding MACR ($38,060,872 increase). Given that no SAMAs are screened in the Phase I analysis based on cost, the extreme flooding mitigation capabilities do not impact the Phase I analysis. E.7.4.2 PHASE II IMPACT If the same changes are made to the credit taken for the extreme flooding mitigation capabilities for each SAMA (i.e., no credit for the current mitigation strategies), the averted cost-risks are altered for those SAMAs that had some impact on the external flooding risk. The following table summarizes the changes to the cost benefit calculations when no credit is taken for the severe flooding mitigation capabilities:
Environmental Report Appendix E  SAMA ANALYSIS Page E-240 Three Mile Island Nuclear Station Unit 1 License Renewal Application Results Summary for the Extreme Flooding Capability Sensitivity SAMA ID Cost of Implement-ation Averted Cost- Risk(Base) Net Value (Base) Averted Cost- Risk(Sensitivity)Net Value (Sensitivity) Change in Cost Effective-ness? SAMA 1 $3,125,000 $986,145 -$2,138,855 $986,143 -$2,138,857 No SAMA 2 $7,300,000 $4,297,001 -$3,002,999 $5,206,254 -$2,093,746 No SAMA 3 $2,450,000 $594,926 -$1,855,074 $601,141 -$1,848,859 No SAMA 5 $3,150,000 $230,163 -$2,919,837 $230,164 -$2,919,836 No SAMA 6 $2,750,000 $398,924 -$2,351,076 $405,139 -$2,344,861 No SAMA 7 $1,000,000 $449,254 -$550,746 $449,256 -$550,744 No SAMA 8 $145,000 $1,234,676 $1,089,676 $1,234,676 $1,089,676 No SAMA 10 $3,800,000 $982,048 -$2,817,952 $982,048 -$2,817,952 No SAMA 11 $4250,000 $16,088,692 $11,838,692 $54,144,650 $49,894,650 No SAMA 12 $50,000 $198,438 $148,438 $198,438 $148,438 No SAMA 13 $950,000 $305,294 -$644,706 $305,294 -$644,706 No SAMA 14 $3,150,000 $603,886 -$2,546,114 $603,888 -$2,546,112 No SAMA 15 $450,000 $199,098 -$250,902 $200,003 -$249,997 No SAMA 16 $1,100,000 $1,592,631 $492,631 $1,592,631 $492,631 No SAMA 17 $950,000 $51,988 -$898,012 $51,988 -$898,012 No SAMA 18 $100,000 $33,260 -$66,740 $40,379 -$59,621 No SAMA 19 $760,000 $3,129,354 $2,369,354 $10,098,967 $9,338,967 No SAMA 20 $3,030,000 $173,974 -$2,856,026 $173,974 -$2,856,026 No SAMA 21 $1,200,000 $1,181,137 -$18,863 $3,908,256 $2,708,256 Yes SAMA 22 $5,000,000 $1,253,768 -$3,746,232 $1,253,770 -$3,746,230 No SAMA 23 $50,000 $30,629 -$19,371 $30,630 -$19,370 No SAMA 24 $8,400,000 $4,416,201 -$3,983,799 $5,325,454 -$3,074,546 No SAMA 25 $6,000,000 1,466,139 -$4,533,861 2,375,392 -$3,624,608 No SAMA 26 $900,000 $369,663 -$530,337 $369,663 -$530,337 No SAMA 27 $575,000 $1,306,819 $731,819 $1,306,819 $731,819 No SAMA 28 $575,000 $47,891 -$527,109 $47,891 -$527,109 No SAMA 29 $800,000 $25,716 -$774,284 $25,716 -$774,284 No SAMA 30 $150,000 $25,716 -$124,284 $25,716 -$124,284 No SAMA 32 $1,700,000 $13,052,022 $11,352,022 $51,109,295 $49,409,295 No SAMA 33 $2,700,000 $9,142,285 $6,442,285 $43,403,546 $40,703,546 No As demonstrated in the table above, of all the SAMAs evaluated, the "cost effectiveness" classification was only changed for SAMA 21. Given that the 95 th percentile PRA results sensitivity presented in S ection E.7.1 also identifi e d this SAMA as potentially cost effec t ive, it can be concluded that the results of the SAMA analysis are not impacted by making pessimistic assumptions related to external flooding risk at TMI-1.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-241 License Renewal Application E.7.5 SENSITIVITY ANALYSIS: IMPACT OF IMPLEMENTING SAMA 32 While the TMI-1 SAMA list is comprised of unique plant enhancements, it is not uncommon for one SAMA to address areas of ri sk that are also addressed by one or more other SAMAs. The implication is t hat implementing a SAMA may impact the net values of the non-implemented SAMAs. D epending on the nature of the SAMAs under consideration, implementation of any given SAMA may result in the reclassification of previously cost beneficial SAMAs as "not cost beneficial". Because SAMA 32 is a potential candidate for implementation at TMI-1, a sensitivity analysis has been performed to evaluate the impac t of its implementation on t he cost benefit analysis. Because implementation of SAMA 32 results in a risk decrease, there is no mechanism that would allow a non-cost beneficial SAMA to become cost beneficial; therefore, this sensitivity analysis only addresses the SAMAs that were classified as cost beneficial in either the base case or the 95 th percentile PRA results sensitivity case. Specifically, these SAMAs include: 2, 7, 8, 11, 12, 15, 16, 19, 21, 23, 24, 26, 27, and 33. E.7.5.1 ANALYSIS PROCESS The intent of this analysis is to quantify the net value of each SAMA assuming that SAMA 32 is already implemented at the site. In order to do this, it was necessary to define the PRA model configuration with SAMA 32 implemented as the new "base case". All model changes made to represent implementation of the other SAMAs were made using the new "base case" as the starting point. This allowed the risk reduction for each SAMA to be measured from the configuration in which SAMA 32 was implemented to the configuration in which SAMA 32 was implemented in conjunction with one additional SAMA. Establishing SAMA 32 as the "base case" required no changes to the internal events model given that SAMA 32 did not impact internal events risk. Consequently, all of the internal events based risk reductions calculated for the cost beneficial SAMAs were unchanged from the original SAMA analysis. The same was true for the non-external flooding external events contributions given that they were directly derived from the internal events results through the use of a multiplier.
Environmental Report Appendix E  SAMA ANALYSIS Page E-242 Three Mile Island Nuclear Station Unit 1 License Renewal Application The external flooding results, however, were impacted by SAMA 32 and it was necessary to review the external flooding frequencies for each of the SAMAs and adjust them to account for the impacts of SAMA 32. For all cases other than for SAMAs 11 and 33, the same quantification strategies described in Section E.6 were used to quantify the external flooding ben e fits of the SAMAs. For SAMAs 11 and 33, additional work was required to define how multiple flood mitigation strategies would impact risk. The following table summarizes the assumptions used to perform the quantifications: Quantification Strategy for Implementation of Multiple Flood Mitigation SAMAs Case Floods >310 Floods 305' to 310' Floods <305' Implementation of SAMA 32 and SAMA 33 SAMA 33 would be the primary action with SAMA 32 as the backup. It is assumed that the failure probability for installing SAMA 33's extended flood gates is the same as the IPEEE value of 5.62E-02 for the 305' to 310' floods (variable HSL1). Implementation of SAMA 32 is then addressed by a human dependence factor. Non-negligible dependence exists between the actions to install the flood gates and to implement SAMA 32, but the dependence is cognitive. As execution would be the majority contributor to the flood HEPs and because the execution and cognitive contributors are not separated for the flood actions, it would be overly conservative to use the dependence factors based on the available HEPs. To simulate the results of a true assessment, a moderate dependence factor (from equation 10-16 of NUREG/CR-1278 (NRC 1983)) is used rather than a high factor, which would likely be appropriate for the cognitive portion of the HEPs. The failure probability for SAMA 32 is, therefore, 0.14. The CDF for this sequence would be calculated as follows:
CDF=2.5E-04*5.62E-02*0.14=1.97E-06 Having higher gates will not impact the installation failure probability significantly. With SAMA 32 implemented, the CDFs
 
should be the same as with SAMA 32 alone.
This SAMA does not impact floods below 305' msl. The CDF should be the same as with SAMA 32 alone. Implementation of SAMA 32 and SAMA 11 The implementation of SAMA 11 would essentially result in a configuration that would supercede that established by SAMA 32. The CDF for this sequence would be calculated by multiplying the flooding frequency by the failure probability of SAMA 11 (2.05E-02): CDF=2.5E-04*2.05E-02=5.12E-06 The failure probabilities for SAMA 11 are dominated by hardware faults and use of a dependence factor is not required for addressing any potential dependence between installation of the flood panels and operation of SAMA 11. The CDF for these sequences should be obtained by multiplying the sequence CDFs by 2.05E-02. The 0.5 multiplier used in the IPEEE for the existing severe flood guidelines is disregarded and excluded from the calculation. These sequences would be improved through implementation of SAMA 11. The original CDF, which is the same as the CDF with SAMA 32 implemented, should be multiplied by 2.05E-
: 02.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-243 License Renewal Application Finally, the 95 th percentile PRA results were used in the quantifications given that they are typically used in the final classification of a SAMA's cost benefit status.
E.7.5.2 RESULTS The following table summarizes the results of the sensitivity analysis. As shown below, only one SAMA that was originally identified as potentially cost beneficial would be reclassified as "not cost beneficial" if SAMA 32 were implemented at the site (SAMA 21). Results Summary for the SAMA 32 Implementation Sensitivity SAMA ID Cost of Implement-ation Averted Cost- Risk (95 th percentile PRA results)Net Value (95 th percentile PRA resultsAverted Cost- Risk (SAMA 32 Implemented, 95 thpercentile PRA results) Net Value (SAMA 32 Implemented, 95 th percentile PRA results) Change in Cost Effective-ness? SAMA 2 $7,300,000 $11,816,753 $4,516,753 $10,016,562 $2,716,562 No SAMA 7 $1,000,000 $1,235,449 $235,449 $1,235,451 $235,451 No SAMA 8 $145,000 $3,395,359 $3,250,359 $3,395,359 $3,250,359 No SAMA 11 $4,250,000 $44,243,903 $39,993,903 $8,339,832 $4,089,832 No SAMA 12 $50,000 $545,705 $495,705 $545,705 $495,705 No SAMA 15 $450,000 $547,520 $97,520 $545,562
$95,562 No SAMA 16 $1,100,000 $4,379,735 $3,279,735 $4,379,738 $3,279,738 No SAMA 19 $760,000 $8,601,659 $7,841,659 $2,528,235 $1,768,235 No SAMA 21 $1,200,000 $3,248,127 $2,048,127 $921,330 -$278,670 Yes SAMA 23 $50,000 $84,230 $34,230 $84,233
$34,233 No SAMA 24 $8,400,000 $12,144,553 $3,744,553 $10,344,362 $1,944,362 No SAMA 26 $900,000 $1,016,573 $116,573 $1,016,573 $116,573 No SAMA 27 $575,000 $3,593,752 $3,018,752 $3,593,752 $3,018,752 No SAMA 33 $2,700,000 $25,141,284 $22,441,284 $2,839,757 $139,757 No For SAMAs 2, 11, 19, 21, 24, and 33, the averted cost-risk reductions were all over $1,000,000. A reduction of this magnitude indicates that a large portion of the risk originally intended to be addressed by these SAMAs was removed by SAMA 32 and implementation may not be appropriate. Final judgements related to these SAMAs would likely have to be made using insights outside of the PRA analysis. The remaining SAMAs (7, 8, 12, 15, 16, 23, 26, and 27) are essentially independent of SAMA 32 and none of their averted cost-risk estimates were impacted by more than 1 percent. No Environmental Report Appendix E  SAMA ANALYSIS Page E-244 Three Mile Island Nuclear Station Unit 1 License Renewal Application changes to the conclusions related to these SAMAs would be expected based on implementation of SAMA 32. E.7.6 SENSITIVITY ANALYSIS: IMPACT OF SEC POP ERROR CORRECTIONS The SECPOP2000 code is used to process population and economic data to serve as input data for the Level 3 PRA code MACCS2 that is used to support SAMA evaluations. The SECPOP2000 code is sponsored by the NRC and is maintained by Sandia National Laboratory. After completion of the TMI SAMA analysis, three SECPOP errors were identified that if uncorrected, result in MACCS2 utilizing incorrect data thereby impacting the SAMA cost benefit calculations. The TMI SAMA evaluation was not impacted by the first SECPOP error described in this disc u ssion (i.e
., Error #1), but the analysis is affected b y the second and third e rrors (Error #2 and Error #3 , respectively). All three e r rors are discussed below for completeness. E.7.6.1 ERROR #1 In May 2007, a formatting error associated with the SECPOP2000 output file option (which generates a text file for use as an input file to MACCS2) was publicized throughout the industry. The error involves the formatting of the columns in the text file resulting in MACCS2 mis-reading the data. Exelon Risk Management was aware of this formatting error well before its publication throughout the industry. For the TMI SAMA analysis, Risk Management had manually corrected the alignment of the SECPOP2000 output for proper reading by MACCS2. As a result, the TMI SAMA evaluation is not impacted by this error. E.7.6.2 ERROR #2 In mid-July 2007, an error associated with the formatting of the 1997 economic database file used by SECPOP2000 was discovered by a MACCS2 industry user. This error was discovered when the user attempted to update the database file with new data, and the SECPOP2000 output did not change. Investigation revealed that a formatting error in the database file resulted in SECPOP2000 processing incorrect economic and land use data (i.e., data is output for the wrong counties). The incorrect county selection results in incorrect data being used in MACCS2, ultimately influencing the SAMA cost benefit calculations. The magnitude of the error's impact on the results is different for each site as it depends on the relative difference between the correct county data and incorrect county data read by SECPOP2000, which varies Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-245 License Renewal Application for each county considered. As a result, a site-specific analysis is required to assess the impact on the cost benefit analysis. E.7.6.3 ERROR #3 In early-August 2007, an additional SECPOP2000 error was identified related to the use of the 1997 economic database file. SECPOP2000 was written to process the county data based on a sequential county numbering system; however, there are gaps in the data file. The first gap appears at county number 955 and any county beyond 955 is handled incorrectly by SECPOP2000. This error was corrected by manipulating the county numbering system in the 1997 economic database file and re-running SECPOP2000.
The nature of Error #3 is similar to Error #2 in that its imp a ct on the c o st benefit a n alysis depends on the relative differences between the correct and incorrect county data. This varies for each county considered and as a result, obtaining an estimate of the impact of the error requires a site specific analysis. E.7.6.4 IMPACT ON TMI MACR Review of t h e TMI SAMA analysis indicates th a t correcting Error #2 and Error #3 results in a measurable change to both of the MACCS2 outputs that are used to quantify the TMI MACR:
* Dose
* Economic cost After addressing the errors, the MACCS2 model was re-quantified and the revised results were used to update the MACR calculation. The following tables provide a summary of the corrected results compared with the base case. The designator "PE23" is used to identify the case in
 
which both Error #2 and Error #3 have been corrected. SECPOP2000 Error Corrections - Internal Events Results Overview CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) Internal Events Results - Base 2.37E-05 32.61 $112,259 Internal Event Results - Post Error Corrections (case PE23) 2.37E-05 32.33 $128,937 Percent Change 0.0% -0.8% 14.8%
Environmental Report Appendix E  SAMA ANALYSIS Page E-246 Three Mile Island Nuclear Station Unit 1 License Renewal Application The following tables provide the release category specific results: Release Category RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Freq. (/yr)PE23 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 Dose-RiskPE23 2.62 9.11 0.92 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.02 0.92 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 OECRPE23 $14,670 $51,039 $3,873 $272 $4 $4 $8 $13 $398 $149 $87 $3,223 $17,071 $3,835 $242 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Freq. (/yr)PE23 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 Dose-RiskPE23 0.00 0.00 0.22 0.04 0.30 0.00 0.99 0.38 6.95 3.42 0.00 0.61 0.00 32.33 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259 OECRPE23 $3 $0 $880 $157 $981 $12 $3,248 $1,260 $22,904 $3,897 $5 $696 $6 $128,937 Based on these results, the revised non-external flooding cost-risk can be calculated using the methodology from Sect i on E.4 and the 2.0 multiplier on the internal events resul t s: SECPOP2000 Error Corrections - Non-External Flooding Cost-Risk PE23 Internal Events Cost-Risk Non-Flood External Events Multiplier Non-Ext. Flooding Cost-Risk $3,514,124 2.0 $7,028,248 Because the Level 3 results are also used in the external flooding evaluation, the impact on the external flooding contribution must also be considered. The following tables summarize the changes to the external flooding results. SECPOP2000 Error Corrections - External Flooding Results Overview CDF (/yr) Dose-Risk (person-rem/yr) OECR ($/yr) External Flooding Results - Base 8.11E-05 177.16 $542,159 External Flooding Results - Post Error Corrections (PE23) 8.11E-05 175.86 $619,814 Percent Change 0.0% -0.7% 14.3%
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-247 License Renewal Application A further breakdown of this information is provided below according to flood sequence. SECPOP2000 Error Corrections - External Flooding Contributions by Flood Sequence Flood Category >310' 305' to 310' sequence A 305' to 310' sequence B 305' to 310' sequence C 305' to 310' sequence D 305' to 310' sequence E 305' to 310' sequence F <305' (uses LOOP RC distribution) Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Freq. (/yr)PE23 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Dose-RiskPE23 131.68 13.02 0.19 1.87 17.81 10.67 0.25 0.37 175.86 Base OECR $405,951 $40,149 $598 $5,767 $54,839 $32,858 $778 $1,219 $542,159
 
OECRPE23 $464,785 $45,968 $678 $6,603 $62,220 $37,281 $882 $1,397 $619,814 Based on these results, the revised external flooding cost-risk can be calculated using the methodology from Sect i on E.4 , which yields $1 6 ,672,271. Finally, the revised TMI MACR is the sum of the External Flooding and non-External Flooding contributors: SECPOP2000 Error Corrections - MACR PE23 Non-External Flooding Cost-Risk PE23 External Flooding  Cost-Risk Total MACR (sum of Ext. Flood and Non-Ext. Flood) $7,028,248 $16,672,271 $23,700,519 Given that the base case MACR was developed by rounding the results of the process documented in Section E.4 to the next highest thousand, t h e same is d one here to obtain a MACR of $23,701,000. This result represents an increase over the base case of 7.3%
(($23,701,000-$22,087,000)/ $22,087,000*100=7.3%). Further investigations revealed that impacts on individual SAMA candidates may differ due to specific release category dependencies (i.e., some release categories may see increases while other release categories see decreases.)  Therefore, changes to the averted cost-risk values for each SAMA candidate can not be readily predicted without a SAMA specific re-quantification, which is ad d ressed in S e ction E.7.6.
: 5. E.7.6.5 IMPACT ON INDIVIDUAL SAMA CALCULATIONS In addition to the impact on the MACR, the SECPOP errors also impacted the averted cost-risks and net values that were calculated for each of the SAMAs. The following table summarizes the impact of all SECPOP2000 error corrections (case PE23) in conjunction with the mean PRA results for the detailed cost-benefit calculations that were performed for the SAMA analysis.
Environmental Report Appendix E  SAMA ANALYSIS Page E-248 Three Mile Island Nuclear Station Unit 1 License Renewal Application Results Summary for SECPOP2000 Corrections (Case PE23, Mean PRA Results) SAMA ID Cost of Implement-ation Averted Cost- Risk(Base) Net Value (Base) Averted Cost- Risk (PE23) Net Value (PE23) Change in Cost Effective-ness? SAMA 1 $3,125,000 $986,145 -$2,138,855 $1,039,690 -$2,085,310 No SAMA 2 $7,300,000 $4,297,001 -$3,002,999 $4,597,411 -$2,702,589 No SAMA 3 $2,450,000 $594,926 -$1,855,074 $624,045 -$1,825,955 No SAMA 5 $3,150,000 $230,163 -$2,919,837 $240,738 -$2,909,262 No SAMA 6 $2,750,000 $398,924 -$2,351,076 $412,415 -$2,337,585 No SAMA 7 $1,000,000 $449,254 -$550,746 $467,015 -$532,985 No SAMA 8 $145,000 $1,234,676 $1,089,676 $1,318,032 $1,173,032 No SAMA 10 $3,800,000 $982,048 -$2,817,952 $1,086,512 -$2,713,488 No SAMA 11 $4250,000 $16,088,692 $11,838,692 $17,237,942 $12,987,942 No SAMA 12 $50,000 $198,438 $148,438 $210,304 $160,304 No SAMA 13 $950,000 $305,294 -$644,706 $333,154 -$616,846 No SAMA 14 $3,150,000 $603,886 -$2,546,114 $630,447 -$2,519,553 No SAMA 15 $450,000 $199,098 -$250,902 $209,632 -$240,368 No SAMA 16 $1,100,000 $1,592,631 $492,631 $1,745,154 $645,154 No SAMA 17 $950,000 $51,988 -$898,012 $55,242 -$894,758 No SAMA 18 $100,000 $33,260 -$66,740 $32,229 -$67,771 No SAMA 19 $760,000 $3,127,876 $2,367,876 $3,415,704 $2,655,704 No SAMA 20 $3,030,000 $173,974 -$2,856,026 $189,934 -$2,840,066 No SAMA 21 $1,200,000 $1,181,137 -$18,863 $1,292,074 $92,074 Yes SAMA 22 $5,000,000 $1,253,768 -$3,746,232 $1,380,631 -$3,619,369 No SAMA 23 $50,000 $30,629 -$19,371 $32,220 -$17,780 No SAMA 24 $8,400,000 $4,416,201 -$3,983,799 $4,730,523 -$3,669,477 No SAMA 25 $6,000,000 1,466,139 -$4,533,861 1,574,565 -$4,425,435 No SAMA 26 $900,000 $369,663 -$530,337 $397,053 -$502,947 No SAMA 27 $575,000 $1,306,819 $731,819 $1,403,645 $828,645 No SAMA 28 $575,000 $47,891 -$527,109 $51,440 -$523,560 No SAMA 29 $800,000 $25,716 -$774,284 $27,621 -$772,379 No SAMA 30 $150,000 $25,716 -$124,284 $27,621 -$122,379 No SAMA 32 $1,700,000 $13,052,022 $11,352,022 $14,000,044 $12,300,044 No SAMA 33 $2,700,000 $9,142,285 $6,442,285 $9,807,683 $7,107,683 No  As demonstrated in the table, the SECPOP2000 error corrections had a relatively small impact on the averted cost-risk estimates and only one SAMA (SAMA 21) that was originally classified as "not cost beneficial" was re-classified as "cost beneficial" based on the use of the corrected input. Given that SAMA 21 was identified as potentially cost beneficial in the 95 th percentile Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-249 License Renewal Application PRA results sensitivity analysis that is documented in Secti o n E.7.1 , this change did not result in the identification of any new potentially cost beneficial SAMAs. In addition to the review of the mean PRA results quantifications, it was necessary to examine how the 95 th percentile PRA results quantifications were impacted given that they were also used to identify potentially cost beneficial SAMAs. The following table provides a summary of the cost benefit calculations using the results of the SECPOP2000 error corrections in conjunction with the 95 th percentile PRA results. In this case, no SAMAs were identified as potentially cost beneficial that were not already identified in original 95 th percentile PRA results sensitivity analysis doc u m ented in Section E.7.1
. Results Summary for SECPOP2000 Corrections (Case PE23, 95 th Percentile PRA Results) SAMA ID Cost of Implement-ation Averted Cost- Risk (Original 95 th  Percentile Results) Net Value (Original 95 th Percentile Results) Averted Cost- Risk(PE23 with 95 th Percentile Results) Net Value (PE23 with 95 th Percentile Results) Change in Cost Effective-ness? SAMA 1 $3,125,000 $2,711,899 -$413,101 $2,859,148 -$265,853 No SAMA 2 $7,300,000 $11,816,753 $4,516,753 $12,642,880 $5,342,880 No SAMA 3 $2,450,000 $1,636,047 -$813,954 $1,716,124 -$733,876 No SAMA 5 $3,150,000 $632,948 -$2,517,052 $662,030 -$2,487,971 No SAMA 6 $2,750,000 $1,097,041 -$1,652,959 $1,134,141 -$1,615,859 No SAMA 7 $1,000,000 $1,235,449 $235,449 $1,284,291 $284,291 No SAMA 8 $145,000 $3,395,359 $3,250,359 $3,624,588 $3,479,588 No SAMA 10 $3,800,000 $2,700,632 -$1,099,368 $2,987,908 -$812,092 No SAMA 11 $4250,000 $44,243,903 $39,993,903 $47,404,341 $43,154,341 No SAMA 12 $50,000 $545,705 $495,705 $578,336 $528,336 No SAMA 13 $950,000 $839,559 -$110,442 $916,174 -$33,827 No SAMA 14 $3,150,000 $1,660,687 -$1,489,314 $1,733,729 -$1,416,271 No SAMA 15 $450,000 $547,520 $97,520 $576,488 $126,488 No SAMA 16 $1,100,000 $4,379,735 $3,279,735 $4,799,174 $3,699,174 No SAMA 17 $950,000 $142,967 -$807,033 $151,916 -$798,085 No SAMA 18 $100,000 $91,465 -$8,535 $88,630 -$11,370 No SAMA 19 $760,000 $8,601,659 $7,841,659 $9,393,186 $8,633,186 No SAMA 20 $3,030,000 $478,429 -$2,551,572 $522,319 -$2,507,682 No SAMA 21 $1,200,000 $3,248,127 $2,048,127 $3,553,204 $2,353,204 No SAMA 22 $5,000,000 $3,447,862 -$1,552,138 $3,796,735 -$1,203,265 No SAMA 23 $50,000 $84,230 $34,230 $88,605
$38,605 No SAMA 24 $8,400,000 $12,144,553 $3,744,553 $13,008,938 $4,608,938 No SAMA 25 $6,000,000 $4,031,882 -$1,968,118 $4,330,055 -$1,669,945 No Environmental Report Appendix E  SAMA ANALYSIS Page E-250 Three Mile Island Nuclear Station Unit 1 License Renewal Application Results Summary for SECPOP2000 Corrections (Case PE23, 95 th Percentile PRA Results) SAMA ID Cost of Implement-ation Averted Cost- Risk (Original 95 th  Percentile Results) Net Value (Original 95 th Percentile Results) Averted Cost- Risk(PE23 with 95 th Percentile Results) Net Value (PE23 with 95 th Percentile Results) Change in Cost Effective-ness? SAMA 26 $900,000 $1,016,573 $116,573 $1,091,894 $191,894 No SAMA 27 $575,000 $3,593,752 $3,018,752 $3,860,024 $3,285,024 No SAMA 28 $575,000 $131,701 -$443,299 $141,459 -$433,541 No SAMA 29 $800,000 $70,719 -$729,281 $75,958 -$724,042 No SAMA 30 $150,000 $70,719 -$79,281 $75,958 -$74,042 No SAMA 32 $1,700,000 $35,893,061 $34,193,061 $38,500,121 $36,800,121 No SAMA 33 $2,700,000 $25,141,284 $22,441,284 $26,971,128 $24,271,128 No Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-251 License Renewal Application E.8 CONCLUSIONS The benefits of revising the operational strategies in place at TMI-1 and/or implementing hardware modifications can be evaluated without the insight from a risk analysis. Use of the PRA in conjunction with cost-benefit analysis methodologies has, however, provided 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 TMI-1, several are cost beneficial based on the methodology applied in this analysis. The baseline Phase II analysis indicates that the following SAMAs are potentially cost beneficial:
* SAMA 8:  Automate Reactor Coolant Pump Trip (on high motor bearing temperature)
* SAMA 11:  Enhance Extreme External Flooding Mitigation Equipment to Address SBO and Loss of Seal Cooling Scenarios
* SAMA 12: Use the DHR System as an Alternate Suction Source for HPI
* SAMA 16: Automate HPI Injection on Low Pressurizer Level
* SAMA 19: Install Battery Backed Hydrogen Igniters or a Passive Hydrogen Ignition System
* SAMA 27: Improve the 480V AC load center welds
* SAMA 32: Pre-stage Severe Flooding Equipment
* SAMA 33: Increase the Flood Protection Height  In addition, when the 95 th percentile PRA results are used in the analysis, the following additional SAMAs are potentially cost beneficial:
* SAMA 2:  Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for Extended EFW Operation
* SAMA 7:  Use Fire Service Water as an Alternate Cooling Source for the ICCW Heat Exchangers Environmental Report Appendix E  SAMA ANALYSIS Page E-252 Three Mile Island Nuclear Station Unit 1 License Renewal Application
* SAMA 15:  Automate Swap to Recirculation Mode
* SAMA 21:  Install Concrete Shields to Block Direct Pathways from the RPV to the Containment Wall and/or Direct Containment Flooding Early in External Flooding Scenarios
* SAMA 23:  Develop Alarm Response Procedures to Direct Operation of RR-V-5 on Low RBEC Flow
* SAMA 24:  Install Damage Resistant, High Temperature RCP Seals with a Diesel Engine as an Alternate Drive for an EFW Pump and a Portable Generator for Level Control Instrumentation
* SAMA 26:  Reroute Cables so that They Do Not Pass Over Ignition Sources in Fire Area CB-FA-2e (West Inverter Room) or Wrap them in Fire Proof Material While the identification of a SAMA as potentially "cost beneficial" indicates that it may be advantageous to implement the SAMAs from a PRA based risk reduction perspective, not all of the SAMAs should be designated as serious candidates for implementation. Some of the SAMAs address the same types of risk such that implementation of a given SAMA would significantly reduce or eliminate the benefit of another SAMA. In addition, there are differences in the level of uncertainty in the PRA bases that support the SAMA cost benefit calculations.
While a particular SAMA may show a large potential risk reduction, it would be inappropriate to justify the expenditure of a large sum of money to address a risk that is likely overstated by pessimistic PRA assumptions or technical limitations.
Table E.8.1 summarizes th e s e considerations for the SAMAs that have been identified as potentially cost beneficial for TMI-1.
In addition, this table provides the following information:
* The implementation cost for the SAMA
* Averted cost-risk (based on the 95 th percentile PRA results),
* Net value (based on the 95 th percentile PRA results), and
* The ratio of the averted cost-risk per dollar of implementation cost (identified as the "dollar per dollar" (DPD) ratio). 
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-253 License Renewal Application E.9 FIGURES TransientsLoss of Nuclear River WaterLarge & Medium LOCASmall & Very Small LOCAISLOCASteam Generator Tube RuptureInternal FloodsLoss of Offsite Power Figure E.2-1 TMI-1 Level 1 CDF Contributions
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-254 Three Mile Island Nuclear Station Unit 1 License Renewal Application Key Equipment Contribution to CDF 0%5%10%15%20%
25%
30%
35%40%45%
50%EDG/SBO DGSwitchyard/O PDCCW R CS NS RW MakeupDecay Heat 125 VD C M ai n SteamFeedw at erAC Electrical Em er. FW Inst. A i r N S CCW D H RW Fi r e S WKey Equipment% CDF Contributio n Figure E.2-2 TMI-1 System Importance Rankings Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-255 License Renewal Application Figure E.2-3 Simplified CET Binning Logic for External Flooding Analysis
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-256 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.10 TABLES TABLE E.2-1 THREE MILE ISLAND PRA MODEL
 
==SUMMARY==
 
Model Revision Date Model Name Internal Events Excluding Internal Flooding (1/yr) Seismic(1/yr) Internal Flooding(1/yr) Total CDF (1/yr) Total LERF (1/yr) Trunc. Limit (1/yr) NotesNov. 1987 Original PRA 4.43E-4 2.70E-6 <1.0E-5 5.5E-4 NA NR 1 Dec. 1992 IPE 4.19E - 4.19E-5 NA NR 2 Dec. 1994 IPEEE Update - 3.21E - NA - 3 Aug. 2000 2000 Update 3.74E 3.0E-06 4.1E-5 3.75E-6 NR 4 Nov. 2001 L2RV2 3.69E 2.56E-6 3.95E-5 2.70E-6 1E-12 5 Jul. 2003 ABSA 3.33E 3.5E-7 3.38E-5 1.39E-6 1E-14 6 Dec. 2004 2004 Rev. 0 3.07E 2.6E-7 3.09E 1E-11 7 Jun. 2005 2004 Rev. 1 3.32E 3.7E-7 3.36E 1E-11 8 June 2007 2004 Rev. 2 2.32E 4.5E-7 2.37E-5 3.02E-06 1E-11 9 Notes: 1. Original PRA for Three Mile Island Unit 1; Truncation limit not reported (NR). All sequences quantified but some are grouped with more severe support states. LERF not computed. 2. 1992 update for the IPE. Control building ventilation failures deleted from model based on physical testing of the system and rooms served. Truncation limit not reported. LERF not computed. 3. IPEEE update. Seismic results not modified since this report. 4. TMI RISKMAN 2000 Update; Level 2 added to model for first time. Truncation limit not reported. Model reflects plant design as of 1998 (see Reference [9]). 5. TMI RISKMAN L2RV2 model:  Level 2 model directly linked with Level 1 sequences. 6. TMI RISKMAN ABSA model:  Revisions in support of responses to peer certification comments. Plant model reflects plant design as of January, 2003 (see Reference [8]). 7. TMI CAFTA 2004, Rev. 0 model:  Initial conversion of ABSA Level 1 model from RISKMAN to CAFTA software. This model was never officially implemented. Level 2 model not revised. 8. TMI CAFTA 2004, Rev. 1 model:  Implementation of various model changes and improvements since Revision 0 (see Reference [4]). Level 2 model not revised. 9. TMI CAFTA 2004, Rev. 2 model:  Implementation of various model changes and improvements since Revision 1 (see Reference [5]). New Level 2 model (see Reference [6]) developed using CAFTA based on February 1993 Level 2 model (see Reference [3]).
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-257 License Renewal Application TABLE E.2-2 CORE MELT BINS BIN # DESCRIPTION 1 Large LOCA, injection failure 2 Large LOCA, early recirculation failure 3 Large LOCA, late recirculation failure 4 Medium LOCA, injection failure 5 Medium LOCA, early recirculation failure 6 Medium LOCA, late recirculation failure 7 Small LOCA, injection failure, steam generators available 8 Small LOCA, recirculation failure, steam generators available 9 Small LOCA, injection failure, steam generators unavailable 10 Small LOCA, early recirculation failure, steam generators unavailable 11 Small LOCA, late recirculation failure, steam generators unavailable 12 Cycling relief valve, injection failure 13 Cycling relief valve, early recirculation failure 14 Cycling relief valve, late recirculation failure 15 Steam generator tube rupture, injection failure, steam generators unavailable 16 Steam generator tube rupture, early recirculation failure, steam generators unavailable 17 Steam generator tube rupture, late recirculation failure, steam generators unavailable 18 Steam generator tube rupture, steam generators available 19 Interfacing-systems LOCA TABLE E.2-3 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1  TRANSIENT CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS GT-004 13 or 14 Depending on time of recirculation failure. GT-005 2 Based on PTS failures assuming to be a part of this core melt bin. GT-006 12 
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-258 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-4 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1  LOOP-SBO CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM  TABLE E.2-2 COMMENTS LOOP-002 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-004 13 or 14 Depending on time of recirculation failure. LOOP-005 12 LOOP-006 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-009 8 LOOP-011 7 LOOP-012 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-014 10 or 11 Depending on time of recirculation failure.
LOOP-015 9 LOOP-016 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-018 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-021 13 or 14 Depending on time of recirculation failure. LOOP-022 12  LOOP-023 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-025 13 or 14 Depending on time of recirculation failure. LOOP-026 12  LOOP-027 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-030 8  LOOP-032 7  LOOP-033 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-036 8  LOOP-038 7  LOOP-039 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-042 8  LOOP-044 7  LOOP-046 10 or 11 Depending on time of recirculation failure.
LOOP-047 9  LOOP-048 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-050 10 or 11 Depending on time of recirculation failure. LOOP-051 9 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-259 License Renewal Application TABLE E.2-4 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1  LOOP-SBO CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM  TABLE E.2-2 COMMENTS LOOP-052 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-055 8  LOOP-057 7  LOOP-058 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-059 7  LOOP-062 8  LOOP-064 7 LOOP-066 10 or 11 Depending on time of recirculation failure. LOOP-067 9  LOOP-068 2 Based on PTS failures assuming to be a part of this core melt bin. LOOP-069 9 TABLE E.2-5 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 VERY SMALL  LOCA CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM  TABLE E.2-2 COMMENTS VSL-004 8 Conservatively binned since depressurization was successful VSL-006 8 VSL-007 2 Based on PTS failures assuming to be a part of this core melt bin.
VSL-011 8 Conservatively binned since depressurization was successful VSL-013 8 
 
VSL-014 2 Based on PTS failures assuming to be a part of this core melt bin. VSL-016 10 or 11 Depending on time of recirculation failure.
VSL-017 2 Based on PTS failures assuming to be a part of this core melt bin. VSL-018 10 VSL-019 9 SSHR is assumed unavailable.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-260 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-6 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 SMALL LOCA CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS SL-002 10 or 11 Depending on time of recirculation failure. SSHR is assumed unavailable.
SL-003 2 Based on PTS failures assuming to be a part of this core melt bin. SL-005 8 SL-006 2 Based on PTS failures assuming to be a part of this core melt bin. SL-008 8 
 
SL-009 2 Based on PTS failures assuming to be a part of this core melt bin. SL-010 10 Failure of early recirculation is assumed. SL-011 9 SSHR is assumed unavailable.
TABLE E.2-7 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 MEDIUM LOCA CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM  TABLE E.2-2 COMMENTS ML-002 5 or 6 Depending on time of recirculation failure. ML-003 4 Injection failure is assumed, even though partial injection was successful. ML-004 4 TABLE E.2-8 CORE MELT BIN ASSIGNMENTS FOR LEVEL  1 LARGE LOCA CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS LL-002 2 or 3 Depending on time of recirculation failure. LL-003 1 Injection failure is assumed, even though partial injection was successful. LL-004 1 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-261 License Renewal Application TABLE E.2-9 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1  SGTR CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS SGTR-004 18  SGTR-006 18  SGTR-010 18  SGTR-012 18 
 
SGTR-013 16 Since SSHR is unavailable, water is assumed to be present in containment due to primary pressure relief.
SGTR-014 16 SSHR is assumed unavailable.
SGTR-015 15 SSHR is assumed unavailable.
TABLE E.2-10 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 STEAMLINE BREAKS UPSTREAM MSIVS CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS SLBI-003 13 or 14 Depending on time of recirculation failure.
SLBI-004 13 Conservative assumption due to failure of pressure relief. SLBI-005 12 SLBI-007 10 or 11 Depending on time of recirculation failure. SSHR is assumed unavailable.
SLBI-008 9 SSHR is assumed unavailable. SLBI-010 13 or 14 Depending on time of recirculation failure.
SLBI-011 13 Conservative assumption due to failure of pressure relief.
SLBI-012 2 Based on PTS failures assuming to be a part of this core melt bin. SLBI-013 12 
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-262 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-11 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 STEAMLINE BREAKS DOWNSTREAM MSIVS CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS SLBO-004 13 or 14 Depending on time of recirculation failure. SLBO-005 13 Conservative assumption due to failure of pressure relief.
SLBO-006 2 Based on PTS failures assuming to be a part of this core melt bin. SLBO-007 12  SLBO-009 10 or 11 Depending on time of recirculation failure. SSHR is assumed unavailable.
SLBO-010 2 Based on PTS failures assuming to be a part of this core melt bin. SLBO-011 9 SSHR is assumed unavailable.
TABLE E.2-12 CORE MELT BIN ASSIGNMENTS FOR LEVEL 1 ATWS CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS ATWS-002 14 Least conservative bin, since core damage may not result for this sequence. ATWS-003 12 ATWS-005 14 Least conservative bin, since core damage may not result for this sequence. ATWS-006 12 ATWS-007 12 Failure of injection is assumed ATWS-008 12 Failure of injection is assumed ATWS-009 1 Injection is assumed ineffective ATWS-010 1 Injection is assumed ineffective
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-263 License Renewal Application TABLE E.2-13 CORE MELT BIN ASSIGNMENTS FOR L EVEL 1 ISLOCA CORE DAMAGE STATES SEQUENCE ID [6] CORE MELT ASSIGNMENTS FROM TABLE E.2-2 COMMENTS ISLOC-001 19 TABLE E.2-14 CONTAINMENT SAFEGUARDS/ISOLATION STATE STATE ID DESCRIPTION A All safeguards available, containment isolated B Fans available, sprays available in injection mode; sprays unavailable in recirculation mode, containment isolated C Fans available; sprays unavailable in injection and recirculation modes, containment isolated D Sprays available in injection and recirculation modes; fans unavailable, containment isolated E Sprays in injection mode available; fans unavailable, sprays unavailable in recirculation mode, containment isolated F No safeguards available, containment isolated G All safeguards available, small isolation failure H Fans available, sprays available in injection mode; sprays unavailable in recirculation mode, small isolation failure I Fans available; sprays unavailable in injection and recirculation modes, small isolation failure J Sprays available in injection and recirculation modes; fans unavailable, small isolation failure K Sprays in injection mode available; fans unavailable, sprays unavailable in recirculation mode, small isolation failure L No safeguards available, small isolation failure M All safeguards available, large isolation failure N Fans available, sprays available in injection mode; sprays unavailable in recirculation mode, large isolation failure O Fans available; sprays unavailable in injection and recirculation modes, large isolation failure P Sprays available in injection and recirculation modes; fans unavailable, large isolation failure Q Sprays in injection mode available; fans unavailable, sprays unavailable in recirculation mode, large isolation failure R No safeguards available, large isolation failure
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-264 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-15 CONTAINMENT EVENT TREE TOP EVENTS EVENT NODE/STATE DESCRIPTION A Containment Bypass Success Containment is available as a barrier to fission product release Failure Containment is not available as a barrier to fission product release (SGTR, ISLOCA) B Containment Isolation Success Containment is isolated Failure Containment is not isolated C Large Isolation Failure Success Isolation failure is small Failure Isolation failure is large D Auxiliary Building Release Success Fission product release is through the Auxiliary Building Failure Fission product release does not go through the Auxiliary Building E Early Containment Failure Success Early containment failure does not take place Failure Early containment failure does occur F Late Containment Failure Success Late containment failure does not take place Failure Late containment failure does occur G Benign Containment Failure Success Containment failure is benign, i.e., leak before break Failure Containment failure is catastrophic H Ex-Vessel Release of Fission Products Success Ex-vessel release is prevented Failure Ex-vessel release is not prevented I Containment Basemat Failure Success Containment failure from basemat melt-through is prevented Failure Containment failure from basemat melt-through occurs J Revaporization Release Success Revaporization release does not take place Failure Revaporization release does occur K Fission Product Scrubbing Success Fission products are scrubbed in containment, steam generator, or Auxiliary Building Failure Fission products are not scrubbed Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-265 License Renewal Application TABLE E.2-16 INDIVIDUAL RELEASE CATEGORY DEFINITIONS RELEASE CATEGORY DEFINITION BASELINE FREQUENCY
(/YR) 1.01 Containment bypass, outside the auxiliary building, with fission product scrubbing, release begins at approximately 4 hrs 4.57E-07 1.02 Containment bypass, outside the auxiliary building, without fission product scrubbing, release begins at approximately 3 hrs 1.59E-06 2.01 Containment bypass, to the auxiliary building, without ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 4 hrs 0.0 2.02 Containment bypass, to the auxiliary building, without ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 3 hrs 1.81E-07 2.03 Containment bypass, to the auxiliary building, wi th ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 4 hrs 0.0 2.04 Containment bypass, to the auxiliary building, wi th ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 3 hrs 1.27E-08 3.01 Large isolation failure, to the auxiliary building, without ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 1.5 hrs 9.07E-11 3.02 Large isolation failure, to the auxiliary building, without ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 1.5 hrs 9.07E-11 3.03 Large isolation failure, to the auxiliary building, with ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 1.5 hrs 1.90E-10 3.04 Large isolation failure, to the auxiliary building, with ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 1.5 hrs 2.88E-10 3.05 Large isolation failure, outside the auxiliary building, without ex-vessel release of fission products, release begins at approximately 1.5 hrs 0.0 3.06 Large isolation failure, outside the auxiliary building, with ex-vessel release of fission products, release begins at approximately 1.5 hrs 0.0 4.01 Small isolation failure, to the auxiliary building, without ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 2.5 hrs 3.90E-08 4.02 Small isolation failure, to the auxiliary building, without ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 2.5 hrs 1.46E-08 Environmental Report Appendix E  SAMA ANALYSIS Page E-266 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-16 INDIVIDUAL RELEASE CATEGORY DEFINITIONS RELEASE CATEGORY DEFINITION BASELINE FREQUENCY
(/YR) 4.03 Small isolation failure, to the auxiliary building, with ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 2.5 hrs 8.54E-09 4.04 Small isolation failure, to the auxiliary building, with ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 2.5 hrs 3.16E-07 4.05 Small isolation failure, to the environment, without ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 2.5 hrs 0.0 4.06 Small isolation failure, to the environment, without ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 2.5 hrs 0.0 4.07 Small isolation failure, to the environment, with ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 2.5 hrs 0.0 4.08 Small isolation failure, to the environment, with ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 2.5 hrs 0.0 5.01 Early containment failure, without ex-vessel fission product release, release begins at approximately 3.25 hrs 7.39E-07 5.02 Early containment failure, with ex-vessel fission product release, release begins at approximately 5.5 hrs  1.66E-07 6.01 Late overpressurization, with catastrophic containment failure, without ex-vessel fission product release, without revaporization, with fission product scrubbing, release begins at approximately 45 hrs 0.0 6.02 Late overpressurization, with catastrophic containment failure, without ex-vessel fission product release, without revaporization, without fission product scrubbing, release begins at approximately 45 hrs 0.0 6.03 Late overpressurization, with catastrophic containment failure, without ex-vessel fission product release, with revaporization, with fission product scrubbing, release begins at approximately 45 hrs 2.20E-08 6.04 Late overpressurization, with catastrophic containment failure, without ex-vessel fission product release, with revaporization, without fission product scrubbing, release begins at approximately 45 hrs 2.36E-10 6.05 Late overpressurization, with catastrophic containment failure, with ex-vessel release of fission products, without revaporization, with fission product scrubbing, release begins at approximately 45 hrs 2.08E-11 6.06 Late overpressurization, with catastrophic containment failure, with ex-vessel release of fission products, without revaporization, without fission product scrubbing, release begins at approximately 45 hrs 0.0 Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-267 License Renewal Application TABLE E.2-16 INDIVIDUAL RELEASE CATEGORY DEFINITIONS RELEASE CATEGORY DEFINITION BASELINE FREQUENCY
(/YR) 6.07 Late overpressurization, with catastrophic containment failure, with ex-vessel release of fission products, with revaporization, with fission product scrubbing, release begins at approximately 45 hrs 8.00E-08 6.08 Late overpressurization, with catastrophic containment failure, with ex-vessel release of fission products, with revaporization, without fission product scrubbing, release begins at approximately 45 hrs 1.43E-08 7.01 Late overpressurization, with benign containment failure, without ex-vessel fission product release, with fission product scrubbing, release begins at approximately 14.5 hrs 2.25E-07 7.02 Late overpressurization, with benign containment failure, without ex-vessel fission product release, without fission product scrubbing, release begins at approximately 14.5 hrs 2.75E-09 7.03 Late overpressurization, with benign containment failure, with ex-vessel release of fission products, with fission product scrubbing, release begins at approximately 14.5 hrs 7.45E-07 7.04 Late overpressurization, with benign containment failure, with ex-vessel release of fission products, without fission product scrubbing, release begins at approximately 14.5 hrs 7.89E-07 8.01 Containment failure from basemat melt-through, with ex-vessel release of fission products, release begins at approximately 36 hrs 3.19E-06 9.01 No containment failure, without ex-vessel fission product release, with fission product scrubbing, release begins at approximately 0.5 hrs 1.32E-05 9.02 No containment failure, without ex-vessel fission product release, without fission product scrubbing, release begins at approximately 2.5 hrs 1.69E-08 9.03 No containment failure, with ex-vessel fission product release, with fission product scrubbing, release begins at approximately 2.5 hrs 2.36E-06 9.04 No containment failure, with ex-vessel fission product release, without fission product scrubbing, release begins at approximately 2.5 hrs 1.91E-08 Environmental Report Appendix E  SAMA ANALYSIS Page E-268 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-17
 
==SUMMARY==
OF REPRESENTATIVE MAAP SEQUENCES FOR TMI-1 SOURCE TERMS MAAP CASE NAME DESCRIPTION EFW SEAL LOCA?SPRAYS ON? FANS ON? TCU HOURS TCD HOURS HLCR HOURS TVF HOURS TCF HOURS TENDHOURS NG FRACTION CSI FRACTIONTM0034 INTACT No cont failure, no exvessel rel, FP scrbbed Y Y Y Y 18.8 26.0 26.7 34.6 NA 48 1.2E-01 4.6E-04 TM0035 BMMT Basemat melt w/o debris cooling Y Y N N 18.7 26.0 26.6 34.7 64.4 48 9.7E-01 8.7E-03 TM0036 LATE - SM Small late containment failure 6 hrs Y N N 8.2 9.0 9.9 16.5 52.1 72 7.0E-01 6.5E-03 TM0037 LATE-LRG Large containment failure Y Y N N 18.8 26.0 26.6 34.8 70.8 72 1.0E+00 6.9E-02 TM0038 EARLY Early containment failure at vessel breach 6 hrs Y N N 8.2 9.3 NA 11.7 11.7 48 1.0E+00 6.0E-02 TM0039 ISO-SM Containment isolation failure -
small N Y N N 0.6 0.8 1.4 6.0 0.0 48 8.3E-01 3.4E-02 TM0040 ISO-LRG Containment isolation failure -
large 6 hrs Y N N 8.5 9.4 10.0 16.0 0.0 48 1.0E+00 2.3E-01 TM0041 ISLOCA .003 ft 2 break N N N N 15.0 15.8 16.8 24.3 NA 72 9.2E-01 1.8E-01 TM0042 SGTR .0066 ft 2 break N N N N 12.7 13.5 16.6 18.3 NA 48 1.0E+00 6.5E-01 Notes to Table E.2-17:
EFW Is EFW available for makeup?
Tcu Time of core uncovery Tcd Time of core damage (max core > 1800F)
Tvf Time of vessel failure Tcf Time of containment failure Tend End time of scenario run NG Noble Gas release CsI CsI release Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-269 License Renewal Application TABLE E.2-18 TMI-1 SOURCE TERM
 
==SUMMARY==
RELEASE CATEGORY INTACT BMMT LATE-SM LATE-LRG EARLY ISO-SM ISO-LRG ISLOCA SGTR MAAP Case ID TM0034 TM0035 TM0036 TM0037 TM0038 TM0039 TM000040 TM0041 TM0042 Run Duration 48 hr 72 hr 72 hr 120 48 hr 48 hr 48 hr 72 hr 48 hr Time after Scram when General Emergency is declared (3) 26 hr 26 hr 9 hr 26 hr 9.3 hr 0.8 hr 9.4 hr 15.8 hr 13.5 hr Fission Product Group:
: 1) Noble                  Total Plume 1 Release Fraction 1.25E-01 3.00E-01 7.00E-01 1.00E+00 1.00E+00 8.30E-01 1.00E+00 9.20E-01 1.00E+00 Start of Plume 1 Release (hr) 26.00 26.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 48.00 64.00 72.00 70.80 11.70 48.00 20.00 20.00 16.00 Total Plume 2 Release Fraction 2  1.00E+00 Start of Plume 2 Release (hr)  64.00 End of Plume 2 Release (hr)  64.00
: 2) CsI                  Total Plume 1 Release Fraction 4.60E-04 8.70E-03 6.50E-03 7.00E-02 6.00E-02 3.40E-02 2.30E-01 1.80E-01 2.00E-02 Start of Plume 1 Release (hr) 26.00 26.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 30.00 50.00 20.00 100.00 11.70 8.00 16.00 25.00 14.00 Total Plume 2 Release Fraction 2                6.50E-01 Start of Plume 2 Release (hr) 34.00 End of Plume 2 Release (hr) 44.00 3) TeO2                  Total Plume 1 Release Fraction 4.60E-04 9.00E-03 9.00E-03 2.00E-02 3.80E-02 1.50E-02 2.00E-01 6.00E-02 1.00E-02 Start of Plume 1 Release (hr) 26.00 26.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 30.00 50.00 20.00 100.00 11.70 8.00 16.00 20.00 14.00 Total Plume 2 Release Fraction 2                4.00E-02 Start of Plume 2 Release (hr) 34.00 End of Plume 2 Release (hr) 44.00 Environmental Report Appendix E  SAMA ANALYSIS Page E-270 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-18 TMI-1 SOURCE TERM
 
==SUMMARY==
RELEASE CATEGORY INTACT BMMT LATE-SM LATE-LRG EARLY ISO-SM ISO-LRG ISLOCA SGTR
: 4) SrO                  Total Plume 1 Release Fraction 7.00E-05 8.50E-04 4.00E-04 5.00E-06 4.50E-03 1.50E-03 1.00E-02 6.00E-03 9.00E-04 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 1.00 12.00 16.00 14.00 End of Plume 1 Release (hr) 32.00 40.00 20.00 70.80 20.00 8.00 20.00 20.00 24.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
: 5) MoO2                  Total Plume 1 Release Fraction 3.50E-04 4.00E-03 2.80E-03 2.00E-05 2.00E-02 2.00E-02 3.50E-02 3.00E-02 6.00E-03 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 32.00 40.00 20.00 70.80 11.70 8.00 16.00 20.00 14.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
: 6) CsOH                  Total Plume 1 Release Fraction 4.50E-04 9.00E-03 5.50E-03 2.00E-02 3.00E-02 1.00E-02 1.50E-01 5.00E-02 2.00E-02 Start of Plume 1 Release (hr) 26.00 26.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 30.00 50.00 20.00 100.00 11.70 8.00 16.00 20.00 14.00 Total Plume 2 Release Fraction 2                9.00E-02 Start of Plume 2 Release (hr) 34.00 End of Plume 2 Release (hr) 44.00 7) BaO                  Total Plume 1 Release Fraction 1.80E-04 3.00E-03 1.00E-03 1.20E-05 5.00E-03 9.00E-03 1.50E-02 2.50E-02 2.00E-03 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 1.00 10.00 16.00 14.00 End of Plume 1 Release (hr) 32.00 40.00 20.00 70.80 20.00 8.00 16.00 20.00 14.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr) 
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-271 License Renewal Application TABLE E.2-18 TMI-1 SOURCE TERM
 
==SUMMARY==
RELEASE CATEGORY INTACT BMMT LATE-SM LATE-LRG EARLY ISO-SM ISO-LRG ISLOCA SGTR 8) La2O3                  Total Plume 1 Release Fraction 2.00E-06 5.50E-05 3.00E-05 5.50E-07 5.50E-04 1.00E-04 9.00E-04 2.50E-04 1.00E-04 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 1.00 14.00 16.00 14.00 End of Plume 1 Release (hr) 32.00 40.00 20.00 70.80 20.00 8.00 20.00 20.00 24.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
: 9) CeO2                  Total Plume 1 Release Fraction 1.00E-05 5.20E-04 5.00E-04 1.00E-05 1.50E-02 1.50E-03 2.00E-02 1.50E-03 2.00E-03 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 4.00 14.00 16.00 14.00 End of Plume 1 Release (hr) 32.00 50.00 20.00 70.80 20.00 10.00 20.00 26.00 24.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
: 10) Sb                  Total Plume 1 Release Fraction 4.00E-04 1.50E-02 8.00E-03 5.00E-02 1.80E-01 5.00E-02 1.50E-01 1.50E-01 7.00E-01 Start of Plume 1 Release (hr) 29.00 30.00 10.00 70.80 11.70 1.00 10.00 16.00 28.00 End of Plume 1 Release (hr) 32.00 40.00 20.00 120.00 20.00 8.00 20.00 20.00 30.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
: 11) Te2                  Total Plume 1 Release Fraction 0.00E+00 1.00E-04 3.00E-05 1.50E-03 2.00E-04 4.00E-03 7.00E-04 9.00E-05 2.00E-04 Start of Plume 1 Release (hr)  30.00 18.00 70.80 11.70 6.00 16.00 30.00 20.00 End of Plume 1 Release (hr)  40.00 20.00 70.80 20.00 16.00 20.00 40.00 24.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr) 
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-272 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-18 TMI-1 SOURCE TERM
 
==SUMMARY==
RELEASE CATEGORY INTACT BMMT LATE-SM LATE-LRG EARLY ISO-SM ISO-LRG ISLOCA SGTR 12) UO2                  Total Plume 1 Release Fraction 0.00E+00 5.00E-06 2.80E-06 1.50E-06 1.20E-04 1.00E-05 2.00E-04 5.00E-06 1.00E-05 Start of Plume 1 Release (hr)  30.00 18.00 70.80 11.70 6.00 16.00 30.00 20.00 End of Plume 1 Release (hr)  50.00 20.00 70.80 20.00 16.00 20.00 40.00 24.00 Total Plume 2 Release Fraction 2                  Start of Plume 2 Release (hr)
End of Plume 2 Release (hr)
Notes to Table E.2-18:
 
(1) Puff releases are denoted in the table by those entries with equivalent start and end times. (2) Plume 2 release fraction is cumulative and includes the initial plume 1 release fraction (3) General Emergency declaration based on time of core damage per Radiological Emergency Plant for TMI, EP-AA-1009 Revision 7
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-273 License Renewal Application TABLE E.2-19 TMI-1 INITIATING EVENT CONTRIBUTIONS TO CDF INITIATOR PROBABILITY %CDF Loss of Offsite Power 7.73E-06 32.6% Transients 5.80E-06 24.5% Small & Very Small LOCA 4.66E-06 19.7% Loss of Nuclear River Water 3.67E-06 15.5% Steam Generator Tube Rupture 9.93E-07 4.2% Internal Floods 4.50E-07 1.9% Large & Medium LOCA 2.06E-07 0.9% ISLOCA 1.80E-07 0.8%  TABLE E.2-20 TMI-1 TOP INITIATING EVENT CONTRIBUTIONS FOR EACH RELEASE CATEGORYRELEASE CATEGORY GROUP RELEASE CATEGORY FREQUENCY (1/YR) PERCENT CONTRIBUTION OF  TOP INITIATING EVENTS 1 2.04E-6 27.5%:  Loss of Instrument Air 25.2%:  "A" Division SGTR 25.2%:  "B" Division SGTR 2 1.93E-7 97.8%:  Interfacing System LOCA 1.0%:  Loss of Offsite Power 0.3%:  Loss of 4160V AC Bus 3 6.60E-10 80.9%:  Loss of Offsite Power 19.1%:  Loss of 4160V AC Bus 4 3.78E-7 87.7%:  Loss of Offsite Power 4.1%:  Loss of 4160V AC Bus 3.0%:  Steam Line Break 5 9.05E-7 35.3%:  Loss of Offsite Power 18.1%:  Loss of Nuclear River Water
 
13.3%:  Very Small LOCA 6 1.17E-7 89.7%:  Loss of Offsite Power 4.3%:  Loss of 4160V AC Bus 2.2%:  Very Small LOCA 7 1.26E-6 90.1%  Loss of Offsite Power 3.8%:  Loss of 4160V AC Bus
 
2.0%:  Very Small LOCA 8 3.19E-6 77.5%:  Loss of Offsite Power
 
6.4%:  Very Small LOCA 4.8%:  Loss of Nuclear River Water 9 1.44E-5 36.5%:  Loss of Offsite Power 18.9%:  Loss of Nuclear River Water
 
10.2%:  Very Small LOCA Environmental Report Appendix E  SAMA ANALYSIS Page E-274 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-21 EXTERNAL FLOODING CDF
 
==SUMMARY==
 
SEQUENCE IDENTIFIER LEVEL 1 SEQUENCE DESCRIPTION FREQUENCY
(/YR)* >310 Feet No detailed core damage progression information is available for these floods in the IPEEE. Based on the available text, successful installation of flood gates would delay the time to equipment damage, but not prevent it (SBO and core damage would still occur). The IPEEE indicates that there should be several hours available between a high water level warning and the onset of flooding, even for hurricane events. This is in addition to the warnings that would exist related to any incoming storm. As a result, it is assumed that the reactor is placed in cold shutdown prior to the onset of site flooding. Core damage ultimately occurs after the failure of the extreme flooding measures. 6.37E-05 305 to 310 feet Sequence "A" Flood event occurs, Offsite power is available, Flood preparations fail given that transition to cold shutdown was successful. 6.30E-06 305 to 310 feet Sequence "B" Flood event occurs, Offsite power is available, Transition to cold shutdown fails, Flood preparations fail given that transition to cold shutdown failed. 6.65E-08 305 to 310 feet Sequence "C" Flood event occurs, Offsite power is unavailable (on-site power OK),
Flood preparations fail given that transition to cold shutdown was successful. 9.05E-07 305 to 310 feet Sequence "D" Flood event occurs, Offsite power is unavailable (on-site power OK),
Transition to cold shutdown fails, Flood preparations fail given that transition to cold shutdown failed. 6.10E-06 305 to 310 feet Sequence "E" Flood event occurs, Offsite power is unavailable, On-site power is unavailable. 3.66E-06 305 to 310 feet Sequence "F" Flood event occurs, Early warning system fails. 8.65E-08 <305 feet A site flood with river levels between 300 and 305 feet occurs only with a dike failure. All safety equipment appears to be contained in buildings that do not have penetrations below the 305 foot level. Offsite power equipment in the switchyard is not damaged until flood levels reach 307 feet, which would imply off-site power is available if the grid is energized (not likely in a hurricane induced event). The CDF estimated in the IPEEE is based on the flooding frequency, the probability of dam failure, and an assumed 0.1 conditional core damage probability. No details are available related to the core damage progression.
2.5E-07 TOTAL  8.11E-05
* Includes credit for current severe flooding guidelines.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-275 License Renewal Application TABLE E.2-22 CET NODE BINNING CHARACTERISTICS CET NODE BINNING CHARACTERISTIC Early Containment Failure is Prevented Used to identify "early" containment failures. Failure of the node denotes an early containment breach has occurred while success indicates that the containment remains intact or fails late. Late Containment Failure is Prevented Failure of the node implies a "late" containment overpressurization failure has occurred. The success path contains both "no containment failure" cases and basemat melt through cases. Containment Failure is Benign For late overpressurization failure cases, success of this node indicates that containment failure results in a "small" release pathway while failure of the node indicates a "large" release pathway has opened. Ex-Vessel Release of Fission Products is Prevented Release of the fission products from the vessel is used to determine whether or not a basemat failure could occur. If the corium is retained in the vessel (success of the node), no basemat failure is possible. Failure of the node requires a subsequent evaluation of the interaction between the corium and the containment floor. Containment Failure From Basemat Melt through is Prevented For those cases in which the fission products are not retained in the vessel, failure of this node implies that the containment basemat fails due to the interaction between the concrete and the corium. Success of the node implies that the containment remains intact.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-276 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.2-23 FLOOD SEQUENCE SOURCE TERM FREQUENCIES FLOOD SEQUENCE SGTR (RC1) ISLOCA (RC2) ISO-LRG (RC3) ISO-SM (RC4) EARLY (RC5) LATE-LRG (RC6) LATE-SM (RC7) BMMT (RC8) INTACT (RC9) >310' Flood Freq.(/yr) 0.00E+00 0.00E+00 0.00E+00 0.00E+00 8.92E
-06 4.93E-06 4.44E-05 1.43E-06 4.05E-06 305' to 310'
 
Sequence A Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 0.00E+00 8.82E
-07 4.88E-07 4.39E-06 1.41E-07 4.01E-07 305' to 310'
 
Sequence B Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 6.65E-08 0.00E
+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 305' to 310'
 
Sequence C Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.27E-07 7.00E-08 6.31E-07 2.03E-8 5.76E-08 305' to 310' Sequence D Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 6.10E-06 0.00E
+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 305' to 310' Sequence E Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 3.66E-06 0.00E
+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 305' to 310' Sequence F Freq. (/yr) 0.00E+00 0.00E+00 0.00E+00 8.65E-08 0.00E
+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 Floods <305' msl 1.68E-07 0.00E+00 0.00E+00 3.32E-09 2.11E-08 0.00E+00 1.09E-08 3.38E-08 1.64E-07
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-277 License Renewal Application TABLE E.2-24 TMI PEER REVIEW
 
==SUMMARY==
OVERALL ASSESSMENT PRA ELEMENT GRADE BASED ON SUB-ELEMENTS Initiating Events (IE) 3 (C) Accident Sequence Evaluation (AS) 3 Thermal Hydraulic Analysis (TH) 2 (C) Systems Analysis (SY) 3 (C) Data Analysis (DA) 3 (C) Human Reliability Analysis (HR) 2 Dependency Analysis (DE) 3 Structural Response (ST) 3 Quantification (QU) 3 Containment Performance Analysis (L2) 2 (C) Maintenance and Update Process (MU) 2 (C) Overall Assessment: The Three Mile Island PRA can be effectively used to support applications involving risk significant determinations supported by deterministic analysis, once the technical issues and recommendations for enhancements that are noted in the element summaries and Fact and Observation Sheets are addressed. When these enhancements are addressed for thermal hydraulics analysis, containment performance analysis, and the maintenance and update process, the current PRA elements are capable of supporting risk-ranking elements. Areas Requiring Enhancement: Significant opportunities for enhancements to support applications involving risk significance determinations were identified for all PRA elements except for Accident Sequence Evaluation, Dependency Analysis, Structural Response, and Level 1 Sequence Quantification. The peer review process for TMI-1 resulted in one 'A', 29 'B', 37 'C', and 14 'D' level F&O findings identified during the review. All 'A' and 'B' F&Os have been closed with the exc e ption of SY
-21. See Section E.2.4. (C):  This identifier is used to denote a grade that is conditional on the resolution of specific review comments; if the comment(s) is/are not resolved, a lower grade would be appropriate.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-278 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.3-1 ESTIMATED POPULATION DISTRIBUTION WITHIN A 50-MILE RADIUS OF THREE MILE ISLAND, YEAR 2034  Sector 0-1 mile (1.00) (1) 1-2 miles (1.78) (1) 2-3 miles (1.00) (1) 3-4 miles (1.14) (1) 4-5 miles (1.22) (1) 5-10 miles (1.33) (1) 10-mile total N 0 228 3110 9798 455 19442 33034 NNE 0 1226 267 684 899 24566 27642 NE 0 1930 465 667 440 4138 7639 ENE 46 228 79 706 1491 3663 6212 E 26 154 51 656 2652 27227 30766 ESE 25 411 230 547 1151 6719 9084 SE 0 1005 77 714 550 5184 7530 SSE 85 389 354 748 448 4606 6630 S 0 0 311 1693 1804 12795 16603 SSW 0 0 625 251 1061 6100 8036 SW 0 2136 567 1991 645 3262 8600 WSW 0 881 199 1785 1276 3569 7710 W 0 3090 448 2491 3593 8835 18456 WNW 0 3273 64 995 1751 17079 23162 NW 0 0 35 0 4158 46674 50867 NNW 0 0 892 1551 4192 32894 39529 Total 182 14950 7774 25277 26565 226752 301500 (1) Ten year radial population growth factor applied to year 2000 census data to develop year 2034 estimate.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-279 License Renewal Application TABLE E.3-2 ESTIMATED POPULATION DISTRIBUTION WITHIN A 50-MILE RADIUS OF THREE MILE ISLAND, YEAR 2034 Sector 0-10 miles 10-20 miles (1.09) (1) 20-30 miles (1.10) (1) 30-40 miles (1.12) (1) 40-50 miles (1.11) (1) 50-mile totalN 33034 16171 11115 12504 66687 139511 NNE 27642 21750 5535 23296 62672 140895 NE 7639 42789 76809 18906 88956 235099 ENE 6212 14482 23100 67848 300119 411762 E 30766 24171 110948 76620 66474 308978 ESE 9084 64191 201929 49553 84839 409596 SE 7530 30257 16040 25237 45943 125006 SSE 6630 69506 22672 26782 145000 270590 S 16603 127880 32539 35911 154350 367283 SSW 8036 54317 71630 37706 87671 259361 SW 8600 13149 31487 45963 36183 135382 WSW 7710 12601 15712 15080 41633 92736 W 18456 32360 56527 25691 33767 166801 WNW 23162 96481 26716 10568 7576 164503 NW 50867 113320 16415 17466 23171 221239 NNW 39529 65264 19006 14046 22665 160510 Total 301500 798690 738178 503178 1267705 3609252 (1) Ten year radial population growth factor applied to year 2000 census data to develop year 2034 estimate.
Environmental Report Appendix E  SAMA ANALYSIS Page E-280 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.3-3 THREE MILE ISLAND MACCS2 CORE INVENTORY ENTRY NUCLIDE (2) THREE MILEISLAND MACCS2 (1) ENTRY NUCLIDE (2) THREE MILE ISLAND MACCS2 (1) 1 Co-58 2.475E+16 31 Te-131m 3.774E+17 2 Co-60 1.890E+16 32 Te-132 3.774E+18 3 Kr-85 3.885E+16 33 I-131 2.645E+18 4 Kr-85m 8.620E+17 34 I-132 3.811E+18 5 Kr-87 6.067E+15 35 I-133 5.549E+18 6 Kr-88 1.702E+18 36 I-134 6.141E+18 7 Rb-86 2.397E+18 37 I-135 5.142E+18 8 Sr-89 2.900E+18 38 Xe-133 5.549E+18 9 Sr-90 3.126E+17 39 Xe-135 2.038E+18 10 Sr-91 3.959E+18 40 Cs-134 6.326E+17 11 Sr-92 4.144E+18 41 Cs-136 1.754E+17 12 Y-90 3.226E+17 42 Cs-137 4.255E+17 13 Y-91 3.548E+18 43 Ba-139 5.105E+18 14 Y-92 4.144E+18 44 Ba-140 4.920E+18 15 Y-93 4.624E+18 45 La-140 4.994E+18 16 Zr-95 4.587E+18 46 La-141 4.661E+18 17 Zr-97 4.661E+18 47 La-142 4.550E+18 18 Nb-95 4.587E+18 48 Ce-141 4.514E+18 19 Mo-99 5.031E+18 49 Ce-143 4.477E+18 20 Tc-99m 4.403E+18 50 Ce-144 3.626E+18 21 Ru-103 4.033E+18 51 Pr-143 4.403E+18 22 Ru-105 2.690E+18 52 Nd-147 1.839E+18 23 Ru-106 1.521E+18 53 Np-239 4.994E+19 24 Rh-105 2.538E+18 54 Pu-238 1.428E+16 25 Sb-127 2.767E+17 55 Pu-239 1.114E+15 26 Sb-129 8.361E+17 56 Pu-240 1.199E+15 27 Te-127 3.677E+16 57 Pu-241 4.957E+17 28 Te-127m 2.741E+17 58 Am-241 7.621E+14 29 Te-129 1.232E+17 59 Cm-242 1.794E+17 30 Te-129m 8.213E+17 60 Cm-244 1.454E+16 1. Core inventory obtained from TMI specific calculation C-1101-900-E-220-178 2. MACCS2 allows up to 60 nuclides input Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-281 License Renewal Application TABLE E.3-4 MACCS2 RELEASE CATEGORIES VS. THREE MILE ISLAND  RELEASE CATEGORIES MACCS2 Release Categories Three Mile Island Release Categories 1-Xe/Kr Noble Gases 2-I CsI 3-Cs CsOH 4-Te TeO2 (Sb (1) & Te2 (2) are included) 5-Sr SrO 6-Ru(Mo) MoO2 (Mo is in Ru MACCS category) 7-La La2O3 8-Ce CeO2 (UO2 (2) are included) 9-Ba BaO (1) The largest release fraction of the TeO2 and Sb category is used (2) These release fractions are typically negligible.
TABLE E.3-5 MACCS2 BASE CASE MEAN RESULTS SOURCE TERM (DESIGNATOR)
RELEASE CATEGORY DOSE (P-SV) DOSE (P-REM) OFFSITE ECONOMIC COST ($) 1 (SGTR) RC1 RC1-02 5.72E+04 5.72E+06 2.78E+10 2 (ISLOCA) RC2 RC2-04 5.05E+04 5.05E+06 1.86E+10 3 (ISO-LRG) RC3 RC3-06 8.91E+04 8.91E+06 3.76E+10 4 (ISO-SM) RC4 RC4-08 2.93E+04 2.93E+06 8.99E+09 5 (EARLY) RC5 RC5-02 6.15E+04 6.15E+06 2.02E+10 6 (LATE-LRG) RC6 RC6-08 2.81E+04 2.81E+06 9.45E+09 7 (LATE-SM) RC7 RC7-04 1.35E+04 1.35E+06 3.82E+09 8 (BMMT) RC8-01 2.22E+04 2.22E+06 6.28E+09 9 (INTACT) RC9 RC9-04 2.67E+03 2.67E+05 2.62E+08
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-282 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.3-6 RELEASE CATEGORY SPECIFIC MACCS2 BASE CASE MEAN RESULTS RELEASE CATEGORY RC1-01 RC1-02 RC2-02 RC2-04 RC3-01 RC3-02 RC3-03 RC3-04 RC4-01 RC4-02 RC4-03 RC4-04 RC5-01 RC5-02 RC6-03 Freq.(/yr)
BASE 4.57E-07 1.59E-06 1.81E-07 1.27E-08 9.07E-119.07E-111.90E-102.88E-103.90E-081.46E-088.54E-09 3.16E-07 7.39E-071.66E-072.20E-08Dose-RiskBASE 2.61 9.09 0.91 0.06 0.00 0.00 0.00 0.00 0.11 0.04 0.03 0.93 4.54 1.02 0.06 OECRBASE $12,705 $44,202 $3,367 $236 $3 $3 $7 $11 $351 $131 $77 $2,841 $14,928 $3,353 $208 Release Category RC6-04 RC6-05 RC6-07 RC6-08 RC7-01 RC7-02 RC7-03 RC7-04 RC8-01 RC9-01 RC9-02 RC9-03 RC9-04 Sum of Annual RiskFreq.(/yr)
BASE 2.36E-10 2.08E-11 8.00E-08 1.43E-08 2.25E-072.75E-097.45E-072.89E-073.19E-061.32E-051.69E-08 2.36E-06 1.91E-082.37E-05 Dose-RiskBASE 0.00 0.00 0.22 0.04 0.30 0.00 1.01 0.39 7.08 3.53 0.00 0.63 0.01 32.61 OECRBASE $2 $0 $756 $135 $860 $11 $2,846 $1,104 $20,033 $3,461 $4 $618 $5 $112,259
 
TABLE E.3-7 EXTERNAL FLOODING BASE CASE MEAN RESULTS Flood Category >310' 305' to 310' Sequence A 305' to 310' Sequence B 305' to 310' Sequence C 305' to 310' Sequence D 305' to 310' Sequence E 305' to 310' Sequence F <305' (uses LOOP RC distribution)
Total External Flood Frequency Base Frequency 6.37E-05 6.30E-06 6.65E-08 9.05E-07 6.10E-06 3.66E-06 8.65E-08 2.50E-07 8.11E-05 Base Dose-Risk 132.75 13.13 0.19 1.89 17.87 10.71 0.25 0.37 177.16 Base OECR 4.06E+05 4.01E+04 5.98E+02 5.77E+03 5.48E+04 3.29E+04 7.78E+02 1.22E+03 $542,159
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-283 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS
%AC 4.48E-02 1.484 LOSS OF OFFSITE POWER The importance of the LOOP initiator flag provides limited information about plant risk given that the LOOP category is broad and includes several different contributors. These contributors are represented by other events in this importance list that better define specific failures that can be investigated to identify means of reducing plant risk. No credible means of reducing the TMI-1 LOOP frequency have been identified. Implementation of the Maintenance Rule is considered to address equipment reliability issues such that no measurable improvement is likely available based on enhancing maintenance practices. It may be possible to improve switchyard work planning and/or practices, but a reliable means of quantifying the impact of these types of changes is not available. No SAMAs suggested. RECOVERY-LOOP-01 4.97E-01 1.216 NONRECOVERY OF OFFSITE POWER This OSP recovery failure event is related to conditions in which only one EDG (potentially the SBO EDG) is available, EFW is successful, but a seal LOCA occurs due to loss of seal cooling. For these cases, auto alignment and load capability for the SBO EDG would allow recovery of emergency AC power in time to prevent seal damage (SAMA 1). Alternatively, damage resistant, high temperature seals could be installed to eliminate most of the seal leakage after loss of cooling and delay core damage long enough to align the SBO EDG or recover OSP. This SAMA also includes the use of a portable 480V AC generator to power a division of battery chargers and maintain MCR control of EFW (SAMA 2).
Environmental Report Appendix E  SAMA ANALYSIS Page E-284 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS
%VSB 2.56E-03 1.177 VERY SMALL BREAK LOCA Multiple failure types contribute, including failures of HPI, DHRW, and DHCCW. The DHRW and DHCCW failures may be eliminated by providing connections from the NSCCW system to the DHR heat exchangers (DH-C-1A/B) to provide emergency heat removal (SAMA 3). Some of the injection failures are caused by division "A" power failures related to "in-series" HPI minimum flow valves MU-V-36 and MU-V-37. These types of failures could be eliminated by powering these two valves from the MCC 1C ESV swing bus (SAMA 4). Alternatively, MU-V-76A and B (and MU-V-77A/B) could be replaced with MOVs to allow rapid alignment of the "C" pump to seal injection (eliminates pump damage from recirc path failures) (SAMA 5). Cross-ties between trains of the DHR related systems would also reduce risk (SAMA 6).
%LNR 3.42E-03 1.177 LOSS OF NUCLEAR RIVER WATER A large majority of the contribution from this event corresponds to the non-recoverable NSRW failures. For many of these contri butors, MU-V-76A and B (and MU-V-77A/B) could be replaced with MOVs to allow rapid alignment of the "C" pump to seal injection (eliminates pump damage from recirc path failures) (SAMA 5). For those contributors where DHRW or DHCCW fail, a hard piped connection to the FSW system could be used to cool the ICCW heat exchangers to provide backup cooling in the event that the normal supply is lost (SAMA 7).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-285 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS LOCA-SIZE-101 7.80E-01 1.164 PROBABILITY THAT RCP SEAL LOCA IS OF VSLOCA CATEGORY Almost 40% of the contributors include hardware failures that would disable NSRW so that the cross-tie from SSRW would not be available. A hard-piped connection from the FSW could be used as a backup supply to the ICCW heat exchangers. This arrangement has the advantage over use of SSRW to NSRW that the integrity of the NSRW system does not need to be confirmed before cooling to the Thermal Barriers can be re-established through alignment of FSW to the ICCW heat exchangers. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action (SAMA 7). For other contributors, MU-V-76A and B (and MU-V-77A/B) could be replaced with MOVs to allow rapid alignment of the "C" pump to seal injection (eliminates pump damage from recirc path failures) (SAMA 5). FLAG-SBOALIGN-1E 5.00E-01 1.103 SBO ALIGNED TO BUS 1E The contributors containing this event lead to RCP seal LOCAs. These events could be mitigated using damage resistant, high temperature seals (SAMA 2). In addition, these events all include an unrecovered failure of the "A" AC division, which leads to failure of the HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4). Even if the SBO EDG functions as designed, the time to align it to an emergency bus is longer that the time to assumed seal damage. If auto alignment and load capability were provided, it would reduce the seal LOCA contribution (SAMA 1). RECOVERY-LOOP-03 8.11E-02 1.101 NONRECOVERY OF OFFSITE POWER This power recovery event is used in cases where no EDGs are available and EFW is initially successful. Installing high temperature, damage resistant seals with a portable generator to power SG level instrumentation for EFW operation would allow long term SBO mitigation (SAMA 2). RARB-STANDBYFLAG 5.00E-01 1.098 BOTH DHRW TRAINS A AND B IN STANDBY The event is associated with loss of DHRW flow events. Use of the NSCCW system to cool the DHR heat exchangers (DH-C-1A/B) would provide alternate heat removal capabilities (SAMA 3).
Environmental Report Appendix E  SAMA ANALYSIS Page E-286 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS NON-RECOV-LNR-IE 2.70E-01 1.096 NON-RECOVERABLE FRACTION OF %LNR EVENTS For many of the non-recoverable loss of NSRW contributors, MU-V-76A and B (and MU-V-77A/B) could be replaced with MOVs to allow rapid alignment of the "C" pump to seal injection (eliminates pump damage from recirc path failures) (SAMA 5). For those contributors where DHRW or DHCCW fail, a hard piped connection to the FSW system could be used to cool the ICCW heat exchangers to provide backup cooling in the event that the normal supply is lost (SAMA 7). GB-EDG-1B---DGFR 2.07E-02 1.095 DIESEL 1B FAILS TO RUN Most of the contributors containing this event lead to RCP seal LOCAs.
These events could be mitigated using damage resistant, high temperature seals (SAMA 2). In addition, these events typically include an unrecovered failure of the "A" AC division, which leads to failure of the HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4). Even if the SBO EDG is available, the time to align it to an emergency bus is longer that the time to assumed seal damage. If auto alignment and load capability were provided, it would re duce the seal LOCA contribution (SAMA 1). GA-EDG-1A---DGFR 2.07E-02 1.081 DIESEL 1A FAILS TO RUN Most of the contributors with EDG "A" failure result in seal LOCAs due to loss of power. A majority of the total is related to REC-LOOP-101 sequences in which EFW is available and the SBO EDG is aligned after seal damage. High temperature, damage resistant seals (SAMA 2) would address most of these cases. Alternatively, providing auto alignment and load capability for the SBO EDG would preclude initial seal damage (SAMA 1). In addition, some of these events include unrecovered failure of the "A" AC division, which leads to failure of all HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-287 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS
%LNS 2.74E-03 1.079 LOSS OF NUCLEAR SERVICES CLOSED COOLING WATER A large portion of the contributors including this event are related to the operator failure to trip the RCPs on loss of cooling. This contribution could be reduced if high temperature sensors on the motor bearing cooling water lines were installed and used to provide automatic trip signals for the pumps (SAMA 8). RECOVERY--LNR-IE 7.30E-01 1.067 RECOVERABLE FRACTION OF %LNR EVENTS Providing a hardpiped connection from the FSW system to the ICCW heat exchangers would provide an alternate cooling source for the ICCW system on loss of NSRW; however, the dependence between the operator action to perform this cross-tie and the one to SSRW would be high or complete and the benefit would be minimal. Enhancing the MU-V-76A/B valves so that they are operable from the MCR would allow the operators to provide a seal injection path for the "C" HPI pump in a timely manner based on different cues (SAMA 5). This option may provide slightly more benefit and would also prevent the seal LOCA that dominates the cutsets that include "RECOVERY--LNR-IE". INHINJ2_MUHHMUOA 1.00E+00 1.067 OPERATOR OPENS CROSS CONNECT
 
VALVES MU-V-76A/B
 
AND STARTS MU-P-1C MU-V-76A and B (and MU-V-77A/B) are the manual HPI swing pump valves, which require local manipulation to align. Providing motor operators to the valves with controls in the MCR would allow for rapid alignment of the "B" HPI pump to either division in accident conditions. This would also allow the "C" pump to be quickly aligned for seal injection (eliminates pump damage from recirc path failures). Provisions for allowing rapid alignment of the valve and pump power sources must also be made in order to make the SAMA fully functional (SAMA 5). JHHOT1-XTIEHEPOA 5.10E-02 1.066 OTHOT1_RCPTHP1OA AND NR-NRSRXTIEHVAOA The contribution from the failure of this JHEP could be reduced if high temperature sensors on the motor bearing cooling water lines were installed and used to provide automatic trip signals for the pumps (SAMA 8). The automation of the RCP trip action would remove the important dependence issue and is considered to be an effective means of addressing this dependent combination for TMI-1.
Environmental Report Appendix E  SAMA ANALYSIS Page E-288 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS DABB1A------BYFD 4.84E-04 1.062 FAILURE OF BATTERY BANK 1A ON DEMAND About 75% of these contributors are LOOP/seal LOCAs with initial EFW success. Installation of the damage resistant, high temperature seals would prevent loss of primary coolant while removing heat with EFW, which would allow for operation out to at least 24 hours and provide recovery opportunities (SAMA 2). Even if the SBO EDG is available, the time to align it to an emergency bus is longer than the time to assumed seal damage. If auto alignment and load capability were provided, it would reduce the seal LOCA contribution (SAMA 1). AV-LOCADV--HCDOA 1.00E+00 1.052 OPERATOR ACTION FAILURE TO LOCALLY OPERATE ADVS ON
 
LOSS OF AIR A large majority of the contributors including AV-LOCADV--HCDOA result from conditions where RCP seal cooling and HPI makeup are lost due to IA valve and power failures. Multiple SAMAs could address these circumstances, including SAMAs 1, 2, 3, 4, 5, 6, and 7; however, TMI-1 has procedures to perform the local ADV operations that are not credited in the PRA model. If these procedures are credited, the RRW of the operator action is reduced below the review threshold. SAMA 9 is used as a surrogate to demonstrate this. GB-EG-Y-1B--DGMM 1.61E-02 1.052 Emergency Diesel Generator 1B in Maintenance Most of the contributors containing this event lead to RCP seal LOCAs. These events could be mitigated using damage resistant, high temperature seals (SAMA 2). In addition, these events typically include an unrecovered failure of the "A" AC division, which leads to failure of the HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4). Even if the SBO EDG is available, the time to align it to an emergency bus is longer that the time to assumed seal damage. If auto alignment and load capability were provided, it would re duce the seal LOCA contribution (SAMA 1).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-289 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS GB1BDG------DGFS 1.13E-02 1.049 DIESEL GENERATOR 1B FAILS TO START Most of the contributors containing this event lead to RCP seal LOCAs. These events could be mitigated using damage resistant, high temperature seals (SAMA 2). In addition, these events typically include an unrecovered failure of the "A" AC division, which leads to failure of the HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4). Even if the SBO EDG is available, the time to align it to an emergency bus is longer that the time to assumed seal damage. If auto alignment and load capability were provided, it would re duce the seal LOCA contribution (SAMA 1).
%SBL 4.50E-04 1.049 SMALL BREAK LOCA There are multiple failure types contributing to the cutsets including this initiating event and no single change other than the installation of an independent injection/heat removal system would address all of these events. As installation of such a system is known not to be cost effective, it is not suggested as a SAMA. A potential change that could reduce some of the risk would be to provide a means of using NSCCW to cool the DHR heat exchangers (DH-C-1A/B) (SAMA 3). Another potential enhancement would be to add inter-train cross-ties to the DHR related systems (DHR, DHRW, and DHCCW) (SAMA 6).
%LGA 1.23E-03 1.047 LOSS OF GA POWER More than half of the contributions including this event are related to operator failure to align the "C" HPI pump for seal injection. If the cross-connect valves were enhance so that they could be controlled from the MCR, this action could be performed in time to prevent seal damage or at least in time to provide an excess flow path for the "C" pump during injection phase to mitigate the loss of the recirc path (SAMA 5). Alternatively, these failures could be addressed by powering valves MU-
 
V-36 and MU-V-37 from MCC 1C ESV to provide a minimum flow path for the "C" HPI pump (SAMA 4).
Environmental Report Appendix E  SAMA ANALYSIS Page E-290 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS RECOVERY-LOOP-04 4.97E-01 1.047 NONRECOVERY OF OFFSITE POWER This recovery term is used in SBO sequences with TD EFW failures. In these cases, there is neither primary nor secondary injection available. An approach similar to what is used to mitigate the extreme external flooding scenarios could be used to address these scenarios. Making it useful for SBO conditions would require permanently installing the portable generator, primary injection pump, and secondary pump so that they could be aligned from the MCR. The submersible pumps would have to be mounted so that the suctions could easily be swapped from a piped water source to the flood water source. This SAMA would also address non-SBO loss of seal cooling cases given the ability to rapidly align alternate seal cooling (SAMA 11). HP-_14A_14BCVAFD 2.03E-04 1.045 HPI Train Fails MOV CCF Op MU-V-14A;14B Failures of the HPI BWST suction path through valves 14A and 14B could be mitigated by proceduralizing the use of the LPI system to operate as the suction path for the HPI pumps in the injection mode (SAMA 12).
Some interlock bypasses may be required. GA-EG-Y-1A--DGMM 1.61E-02 1.043 Emergency Diesel Generator 1A in Maintenance Most of the contributors with EDG "A" failures result in seal LOCAs due to loss of power. A majority of the total is related to REC-LOOP-101 sequences in which EFW is available and the SBO EDG is aligned after seal damage. High temperature, damage resistant seals (SAMA 2) would address most of these cases . Alternatively, providing the ability to rapidly align the SBO EDG would preclude initial seal damage (SAMA 1). In addition, some of these events include unrecovered failure of the "A" AC division, which leads to failure of all HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-291 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS GA1ADG------DGFS 1.13E-02 1.042 DIESEL GENERATOR 1A FAILS TO START Most of the contributors with EDG "A" failures result in seal LOCAs due to loss of power. A majority of the total is related to REC-LOOP-101 sequences in which EFW is available and the SBO EDG is aligned after seal damage. High temperature, damage resistant seals (SAMA 2) would address most of these cases . Alternatively, providing the ability to rapidly align the SBO EDG would preclude initial seal damage (SAMA 1). In addition, some of these events include unrecovered failure of the "A" AC division, which leads to failure of all HPI makeup pumps for this initiator due to loss of the "C" HPI pump minimum flow path. These failures could be addressed by powering valves MU-V-36 and MU-V-37 from MCC 1C ESV (SAMA 4).
%SLT 4.22E-03 1.038 STEAM LINE BREAK IN TURBINE BUILDING A large the contributor to TB steam line break scenarios is the failure of the operators to start the IA compressors on emergency power after a low voltage trip in conjunction with an ESAS.
If the IA system logic were altered to automatically load the IA-P-1A/B compressors when power is restored after an ESAS, this would reduce the probability that IA would not be available (SAMA 13). GA-1A1BSBO-CDGFR 1.53E-04 1.037 EDG CCF Run DG-1A;DG-1B;DG-SBO The primary contribution from this event comes from SBO with initial success of the TD EFW pump. Installing the high temperature, damage resistant seals will prevent a signific ant seal LOCA and using a portable 480V AC generator to power a battery charger would allow long term operation of the TD EFW pump (SAMA 2).
Environmental Report Appendix E  SAMA ANALYSIS Page E-292 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS EFEFP1------P7FR 5.06E-02 1.036 TURBINE-DRIVEN PUMP EF-P-1 FAILS TO RUN Most of the contribution related to this event comes from SBO sequences. In these cases, there is neither primary nor secondary injection available. An approach similar to what is used to mitigate the extreme external flooding scenarios could be used to address these scenarios. Making it useful for SBO conditions would require permanently installing the portable generator, primary injection pump, and secondary pump so that they could be aligned from the MCR. The submersible pumps would have to be mounted so that the suctions could easily be swapped from a piped water source to the flood water source. This SAMA would also address non-SBO loss of seal cooling cases given the ability to rapidly align alternate seal cooling (SAMA 11). HP-MU-P-1B--P2MM 7.46E-03 1.036 Makeup Pump (Operating) 1B in Maintenance Many of the contributors related to this event could be eliminated if the HPI pump cooling supply valves were replaced with MOVs controllable from the MCR. This would allow rapid alignment of an alternate cooling source to available pumps in the event that the normal supply is lost (SAMA 14). JHHHL1AHSR2HEPOA 2.00E-04 1.035 DLHHL1A----HVHOA AND SAHSR2-----
HSROA This joint human error pr obability includes operat or failure to perform swap to recirculation mode and failure to open the drop line. A potential change that could reduce some of the risk would be to automate the swap to recirculation mode when the BWST has been depleted (SAMA 15). FLAG-SBOALIGN-1D 5.00E-01 1.034 SBO ALIGNED TO BUS 1D The contributors containing this event lead to RCP seal LOCAs. If auto alignment and load capability were provided, it would reduce the seal LOCA contribution (SAMA 1). Alternatively, these events could be mitigated using damage resistant, high temperature seals (SAMA 2). HA-P-1AP-1BCP2FS 1.50E-04 1.034 DH Clsd Cool Stdby Pmp CCF Strt P2-1A;1B A majority of the CCF DHCCW pump failures are important because they fail the heat sink for DHR. Failures of the DHCCW pumps may be mitigated by providing a connection from the NSCCW system to the DHR (DH-C-1A/B) heat exchangers to provide emergency heat removal (SAMA
: 3)
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-293 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS
%LAIR 5.23E-03 1.033 LOSS OF AIR INITIATING EVENT A primary contributor related to this in itiating event is related to the failure of operators to operate the EFW flow control valves after loss of air and a dependent operator failure to initiate HPI. Providing logic to auto-start HPI on low pressurizer level would reduce the risk of this scenario (SAMA 16).
In addition, a connection to the plant Service Air system exists that is not currently credited in the model. Use of this system to recover IA is possible if the integrity of the IA system is not compromised by the IE. Crediting this cross-tie would also reduce the importance of this IE. Finally, a significant contributor is event "AV-LOCADV--HCDOA", which is addressed above. JHHOTHMRXTIHEPOA 3.10E-03 1.033 OTHOT1_RCPTHP1OA; MRHMR1-----HMUOA; NR-NRSRXTIEHVAOA This JHEP is important for cases where %LNR has failed thermal barrier cooling, contributed to loss of seal injection, created a small LOCA via loss of RCP bearing cooling, and failed the remaining HPI source. These types of scenarios would be reduced in frequency by automating RCP trip on high motor bearing coolant temperature (SAMA 8). HADC-V-2A---VCFT 3.00E-03 1.031 DC-V2A FAILS TO REMAIN OPEN Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment.
Environmental Report Appendix E  SAMA ANALYSIS Page E-294 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS HADC-V-65A--VCFT 3.00E-03 1.031 DC-V65A TRANSFERS TO  DIFFERENT STATEProviding cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment. RB-RUNNING--FLAG 2.50E-01 1.026 DHRW TRAIN B RUNNING AND TRAIN A IN STANDBY Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). HA-DC-P-1A--P1MM 2.84E-03 1.026 Decay Heat Closed Cycle Cooling Water Pump 1A in MaintenanceProviding cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-295 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS LOCA-SIZE-100 2.20E-01 1.026 PROBABILITY THAT RCP SEAL LOCA IS OF SLOCA CATEGORY About 50% of the contributors including LOCA-SIZE-100 result from the failure of NSRW to cool ICCW and to supply cooling to the running makeup pump in conjunction with failures that eliminate the remaining trains of seal injection. A hard-piped connection from the FSW could be used as a backup supply to the ICCW heat exchangers and maintain seal cooling. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action (SAMA 7). OP-OPB-CONDITION 3.00E-01 1.026 POWER SUPPLY UNAVAILABLE GIVEN A TURBINE BYPASS SIGNAL This event represents the probability that non-emergency electrical power will be lost due to damage from a steam line break in the turbine building. A large the contributor to these scenarios is the failure of the operators to start the IA compressors on emergency power after a low voltage trip in conjunction with an ESAS. If the IA system logic were altered to automatically load the IA-P-1A/B compressors when power is restored after an ESAS, this would reduce t he probability that IA would not be available (SAMA 13). NRHNS8A----HP1OA 5.37E-01 1.025 OPERATOR FAILS TO ISOLATE FAILED RW
 
PUMP (POWER UNAVAILABLE) A large majority of the contributors including event NRHNS8A----HP1OA include the operator failure to open valves MU-V-76A/B to allow seal injection with the "C" HPI pump. MU-V-76A and B (and MU-V-77A/B) are the manual HPI swing pump valves, which require local manipulation to align. Providing motor operators to the valves with controls in the MCR would allow for rapid alignment of the "B" HPI pump to either division in accident conditions. This would also allow the "C" pump to be quickly aligned for seal injection (eliminates pump damage from recirc path failures). Provisions for allowing rapid alignment of the valve and pump power sources must also be made in order to make the SAMA fully functional (SAMA 5).
Environmental Report Appendix E  SAMA ANALYSIS Page E-296 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS HADC-P-1A---P2FS 2.46E-03 1.025 DHCCW PUMP DC-P1A FAILS TO START Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment. JHHEF1-HBW1HEPOA 1.00E-04 1.024 EFHEF1_OPERH2HOA AND BWHBW1-----
HP2OA Providing logic to auto-start HPI on low pressurizer level would reduce the risk of the scenarios including operator failures to initiate HPI (SAMA 16).
It should be noted, however, that a connection to the plant Service Air system exists that is not currently credited in the model. Use of this system to recover IA is possible if the integrity of the IA system is not compromised by the IE. INMU-P-1C--HMUOA 1.00E+00 1.024 OPERATOR FAILURE TO ALIGN AND START MU-P-1C Alignment of the "C" HPI pump for seal injection cannot be accomplished in time to prevent RCP seal damage due to the local, manual valve actions required to get the cooling flow aligned. Providing the ability to perform the alignment rapidly from the MCR would allow this action to be taken in the required time frame (SAMA 5).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-297 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS NRNR-V-20A--VPFD 1.35E-03 1.024 CHECK VALVE NR-V20A FAILS TO RESEAT These events are tied to loss of NSRW due to back flow of a tripped pump. The main contributors include loss of div "A" power so that only HPI pump "C" is available for seal injection/makeup. Because MU-V-76A/B require local operation to align the "C" pump for seal injection, time is not available to perform the alignment before a seal LOCA occurs and the loss of "A" power fails the min flow recirc path, so all HPI will be lost. Providing motor operators to the valves with controls in the MCR would allow the "C" pump to be quickly aligned for seal injection (eliminates pump damage from recirc path failures). Provisions for allowing rapid alignment of the valve and pump power sources must also be made in order to make the SAMA fully functional (SAMA 5). Alternatively, FSW could be used as an alternate cooling medium for the ICCW heat sinks to maintain thermal barrier cooling (SAMA 7). TH-HPIOFF--HP2OA 1.00E+00 1.024 OPERATOR FAILS TO SECURE ALL MU/HPI PUMPS TO PREVENT OVERCOOLING This action is primarily associated with steam line breaks. Inclusion of logic to auto isolate the steam generators on high steam line flow would reduce the isolation failure (SAMA 17). NR-NRSRXTIEHVAOA 1.00E-01 1.024 OPERATOR FAILS TO PERFORM CROSS-TIE
 
IN TIME TO PREVENT LOSS OF RCP SEAL COOLING Many contributors including event "NR-NRSRXTIEHVAOA" also include failure to locally operate MU-V-76A and B (and MU-V-77A/B) to align the "C" HPI pump for seal injection. Providing motor operators to the valves with controls in the MCR would allow the "C" pump to be quickly aligned for seal injection (eliminates pump damage from recirc path failures). Provisions for allowing rapid alignment of the valve and pump power sources must also be made in order to make the SAMA fully functional (SAMA 5). In addition, other contributors include failure of both NSRW and DHCCW. In these cases, FSW could be used as an alternate cooling medium for the ICCW heat sinks to maintain thermal barrier cooling (SAMA 7). JHAHCD4RE27HEPOA 9.17E-05 1.023 AVHCD4_FF--HCDOA AND BWST-HRE27-
 
HTKOA Automating BWST refill would effectively eliminate this JHEP and provide a reliable means of maintaining level in the BWST (SAMA 10).
Environmental Report Appendix E  SAMA ANALYSIS Page E-298 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS JHHNS6-HOT1HEPOA 3.00E-02 1.023 NSHNS6-----HHXOA AND OTHOT1_RCPTHP1OA Automating Reactor Coolant Pump Trip on high motor bearing coolant temperature would eliminate this JHEP and reduce the probability of seal failures (SAMA 8). HBDC-V-2B---VCFT 3.00E-03 1.023 DC-V2B FAILS TO REMAIN OPEN Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). In addition, some scenarios could be mitigated by enhancing the SBO DG so that it could be rapidly aligned to either division. This would benefit the cases where the SBO EDG is aligned to a particular division only to flow up with an equipment failure specific to that division (SAMA 1). HBDC-V-65B--VCFT 3.00E-03 1.023 DC-V65B TRANSFERS TO  DIFFERENT STATEProviding cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment. OP230KV-----OGFD 2.40E-03 1.022 LOSS OF 230KV TO AUX XFRMR 1A AND 1B Many contributors to consequential LOOP events could be addressed by installing high temperature, damage proof seals in conjunction with a 480V AC generator to support continued EFW operation from the MCR (SAMA 2). Other contributors would benefit from changing the IA system logic so that it automatically reloads after power is restored (SAMA 13).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-299 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS RA-RUNNING--FLAG 2.50E-01 1.022 DHRW TRAIN A RUNNING Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3).
%TRIB 2.86E-03 1.021 INITIATING EVENT FOR SGTR ON OTSG B Over 80% of the contribution from the cutsets including this initiating event include operator failures to refill the BWST. Automating refill of the BWST is a potential means of improving the reliability of the refill function (SAMA 10). %TRIA 2.86E-03 1.021 INITIATING EVENT FOR SGTR ON OTSG A Over 80% of the contribution from the cutsets including this initiating event include operator failures to refill the BWST. Automating refill of the BWST is a potential means of improving the reliability of the refill function (SAMA 10). HB-DC-P-1B--P1MM 2.84E-03 1.021 Decay Heat Closed Cycle Cooling Water Pump 1B in MaintenanceProviding cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment.
Environmental Report Appendix E  SAMA ANALYSIS Page E-300 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS RADR-V-1A---VAFD 3.28E-03 1.021 DR-V-1A FAILS TO OPEN ON DEMAND Failure of DR-V-1A contributes to both long term recirculation failures and LOOP related seal LOCAs. Installing a cross-connect from NSCCW to the DHR heat exchangers would provide an alternate means of removing decay heat for many of the loss of DHR cases (SAMA 3). Alternatively, adding cross-ties between the DHR systems would allow the operators to establish DHR in cases where opposite trains of the DHR systems are failed for different reasons (SAMA 6). The LOOP induced seal LOCAs typically occur because the SBO EDG cannot be aligned in time to provide power for seal cooling. Enhancing the SBO EDG with auto alignment and load capability would reduce these contributions (SAMA 1).EF-CCFEFW-LETHAL 4.25E-04 1.02 LETHAL SHOCK TO THE EFW SYSTEM DUE TO COMMON CAUSE FAILURES There are multiple contributors to cutsets including lethal EFW CCF, but about 40% are related to operator failure to manually initiate HPI. Automating HPI initiation on low level would reduce the reliance on operator action to perform this function (SAMA 16). GSHEO1A----HDGOA 2.66E-02 1.019 OPERATOR FAILS TO STARTSBODG Over 90% of the cutsets including this event are SBO sequences and 65% are SBOs in which EFW is initially available. Auto start and load capability for the SBO EDG would essent ially eliminate the contribution of these failures (SAMA 1). The scenarios with EFW available could be addressed by installing high temperature, damage resistant seals that would prevent seal LOCAs (SAMA 2). HBDC-P-1B---P2FS 2.46E-03 1.018 DHCCW PUMP DC-P1A FAILS TO START Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-301 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS
%LGB 1.23E-03 1.018 LOSS OF GB POWER Many of the %LGB events are coupled with what are assumed to be non-recoverable electrical failures of the "A" division that fail "A" HPI. The result is a seal LOCA with no makeup capability. A potential mitigation method would be to permanently mount the extreme flooding equipment so that seal injection and secondary side cooling are available in SBO equivalent conditions (SAMA 11). MRHMR1-----HMUOA 1.03E-02 1.018 OPERATOR FAILS TO RECOGNIZE AND ESTABLISH MIN FLOW RECIRC PATH A large majority of the contributors containing this event are combined with the "INHINJ2_MUHHMUOA" operator action to cross-connect the "C" HPI pump for seal injection. Either "MRHMR1-----HMUOA" or "INHINJ2_MUHHMUOA" would provide a minimum flow path for the "C" pump, but the alignment of the pump for seal injection is a more visible and familiar cue that would prevent damage to the pump. Replacing the MU-V-76A/B valves (and 77A/B for easy swap of the "B" pump) would allow the operator to perform the alignment of the "C" pump in a timely manner and reduce the contribution from these scenarios (SAMA 5). JHHAMHEFHBWHEPO
 
A 2.40E-04 1.017 JHHAM2-HEF1HEPOA AND BWHBW1-----
HP2OA Nearly 80% of the contribution including this cutset is related to a steamline break that causes a trip of the off-site power source and subsequently requires the re-loading of IA onto emergency power. If the IA logic were modified to automatically re-load IA once emergency power is established, the requirement for the operator action would be removed (SAMA 13). HADC-P-1A---P2FR 1.63E-03 1.016 DHCCW PUMP DC-P1A FAILS DURING
 
OPERATION Failure of HADC-P-1A---P2FR contributes to both long term recirculation failures and LOOP related seal LOCAs. Installing cross-connects from NSCCW to the DHR heat exchangers would provide an alternate means of removing decay heat for many of the loss of DHR cases (SAMA 3). Alternatively, adding cross-ties between the DHR systems would allow the operators to establish DHR in cases where opposite trains of the DHR systems are failed for different reasons (SAMA 6). The LOOP induced seal LOCAs typically occur because the SBO EDG cannot be aligned in time to provide power for seal cooling. Enhancing the SBO EDG with auto start and load capability would reduce these contributions (SAMA 1).
Environmental Report Appendix E  SAMA ANALYSIS Page E-302 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS GA-1A-1B---CDGFR 2.31E-04 1.016 EDG CCF Run DG-1A;DG-1B Many of the contributors including this event could be mitigated by enhancing the SBO EDG auto start and load capability so that it can restore seal cooling in time to prevent a seal LOCA (SAMA 1). In other cases, the SBO EDG is failed and would not be available. In these cases, replacing the RCP seals with high temperature, damage resistant seals would allow the operators to maintain RCS integrity and remove heat with the EFW system. Typically, a portable 480V AC generator would be required to provide instrument and control power for EFW to improve the reliability of EFW operation (SAMA 2). DABATTCHGR-HBCOA 1.00E-01 1.016 HEP FOR FAILURE TO ALIGN SPARE CHARGER 1E OR 1F This action is proceduralized at the plant, but the time requirements and reliability of the action could be improved by providing controls in the MCR (SAMA 18). RADR-V-1B---VAFD 3.28E-03 1.015 DR-V-1B FAILS TO OPEN ON DEMAND Failure of DR-V-1B contributes to both long term recirculation failures and LOOP related seal LOCAs. Installing cross-connects from NSCCW to the DHR heat exchangers would provide an alternate means of removing decay heat for many of the loss of DHR cases (SAMA 3). Alternatively, adding cross-ties between the DHR systems would allow the operators to establish DHR in cases where opposite trains of the DHR systems are failed for different reasons (SAMA 6). The LOOP induced seal LOCAs typically occur because the SBO EDG cannot be aligned in time to provide power for seal cooling. Enhancing the SBO EDG with the capability to auto start and load would reduce these contributions (SAMA 1).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-303 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS GS-SBODG----DGFR 2.07E-02 1.015 SBO DIESEL FAILS TO RUN More than half of the contributions including this event are related to SBO cases in which the EFW system is available. For these cases, installing damage resistant, high temperature seals could be installed to eliminate most of the seal leakage after loss of cooling and delay core damage long enough to align the SBO EDG or recover OSP. This SAMA also includes the use of a portable 480V AC generator to power a division of battery chargers and maintain MCR control of EFW (SAMA 2). An additional 25%
of the cases are related to SBO events where the EFW system fails. The result is a seal LOCA with no makeup capability. A potential mitigation method would be to permanently mount the extreme flooding equipment so that seal injection and secondary side cooling are available in SBO equivalent conditions (SAMA 11). HA-P-1AP-1BCP2FR 6.12E-05 1.013 DH Clsd Cool Stndby Pmp CCF Run P2-1A;1BA majority of the CCF DHCCW pump failures are important because they fail the heat sink for DHR. Failures of the DHCCW pumps may be mitigated by providing connections from the NSCCW system to the DHR (DH-C-1A/B) heat exchangers to provide emergency heat removal (SAMA
: 3) AMSC-V-52B--VCFD 6.38E-03 1.013 AIR OPERATED VALVE SC-V-52B FAILS TO OPEN/D About 70% of the contributors including this event also include the event "AV-LOCADV--HCDOA", which is cons ervatively modeled in the TMI-1 PRA model. SAMA 9 demonstrates that when appropriate credit is taken for this action, the RRW is reduced below the SAMA review cutoff level. AMSC-V-58---VCFD 6.38E-03 1.013 F.S. COOLING IA-P1A SC-V-58/D About 70% of the contributors including this event also include the event "AMSC-V-58---VCFD", which is conser vatively modeled in the TMI-1 PRA model. SAMA 9 demonstrates that when appropriate credit is taken for this action, the RRW is reduced below the SAMA review cutoff level. FLAG----NRNORMAB 3.23E-01 1.013 FRACTION THAT NR PUMPS A AND B ARE NORMALLY RUNNING A large majority of the contributors including this event also include the event "INHINJ2_MUHHMUOA", which represents the failure of the operators to align the "C" HPI pump for seal injection. Enhancing the MU-V-76A/B valves so that they are operable from the MCR would allow the operators to provide a seal injection path for the "C" HPI pump in a timely manner based on different cues (SAMA 5).
Environmental Report Appendix E  SAMA ANALYSIS Page E-304 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS JHHMR1-XTIEHEPOA 2.30E-03 1.013 MRHMR1-----HMUOA AND NR-NRSRXTIEHVAOA A large majority of the contributors containing this event are combined with the "INHINJ2_MUHHMUOA" operator action to cross-connect the "C" HPI pump for seal injection. Either "MRHMR1-----HMUOA" or "INHINJ2_MUHHMUOA" would provide a minimum flow path for the "C" pump, but the alignment of the pump for seal injection is a more visible and familiar cue that would prevent damage to the pump. Replacing the MU-V-76A/B valves (and 77A/B for easy swap of the "B" pump) would allow the operator to perform the alignment of the "C" pump in a timely manner and reduce the contribution from these scenarios (SAMA 5). RADR-P-1A---P5FR 1.51E-03 1.013 FAILURE OF DECAY HEAT RIVER WATER PUMP A (DR-P1A) TO RUN Failure of DR-P1A contributes to both long term recirculation failures and LOOP related seal LOCAs. Installing cross-connects from NSCCW to the DHR heat exchangers would provide an alternate means of removing decay heat for many of the loss of DHR cases (SAMA 3). Alternatively, adding cross-ties between the DHR systems would allow the operators to establish DHR in cases where opposite trains of the DHR systems are failed for different reasons (SAMA 6). The LOOP induced seal LOCAs typically occur because the SBO EDG cannot be aligned in time to provide power for seal cooling. Enhancing the SBO EDG with the capability to auto start and load would reduce these contributions (SAMA 1). RA-V-1AV-1BCVAFD 1.34E-04 1.012 DHRW MOV CCF Operate on Demand V-
 
1A;1B A large majority of the contributors including this event are related to failure of the DHRW system to provide long term heat removal. These contributors could be addressed by providing an alternate method of cooling the DHR heat exchangers (DH-C-1A/B) with NSCCW (SAMA 3).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-305 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS GAEDG-STARTCDGFS 5.25E-05 1.012 EDG Fail to Start CCF DG-All 3 About 69% of the contributions including this event are related to SBO cases in which the EFW system is available. For these cases, installing damage resistant, high temperature seals could be installed to eliminate most of the seal leakage after loss of cooling and delay core damage long enough to align the SBO EDG or recover OSP. This SAMA also includes the use of a portable 480V AC generator to power a division of battery chargers and maintain MCR control of EFW (SAMA 2). An additional 29%
of the cases are related to SBO events where the EFW system fails. The result is a seal LOCA with no makeup capability. A potential mitigation method would be to permanently mount the extreme flooding equipment so that seal injection and secondary side cooling are available in SBO equivalent conditions (SAMA 11). HBDC-P-1B---P2FR 1.63E-03 1.012 DHCCW PUMP DC-P1B FAILS DURING OPERATION Providing cross-ties between the DHR cooling water systems (DHRW, DHCCW, and DHR) would provide a means of restoring cooling to the HPI pumps and the DHR heat exchangers in many cases (SAMA 6). In addition, some contributors could be addressed by providing an alternate means of flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). It should be noted that while the ability to rapidly transfer the SBO EDG to the alternate division of power exists, no credit is taken for this capability in the model. As a result, equipment failures after SBO EDG alignment are not recovered while there is a chance that the SBO EDG could be aligned to the opposite division to support use of potentially available equipment. HPMU-P-1A---P2FS 2.46E-03 1.012 MAKEUP PUMP A FAILS TO START Over half of the contribution from this event is related to seal LOCAs in which NSRW cooling to ICCW is lost. Providing an alternate means of cooling the ICCW heat exchangers would prevent the seal LOCAs in these sequences. FSW could be used as a backup cooling source for the ICCW heat exchangers. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action (SAMA 7).
Environmental Report Appendix E  SAMA ANALYSIS Page E-306 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-1 LEVEL 1 IMPORTANCE LIST REVIEW EVENT NAME PROB-ABILITY RED W DESCRIPTION POTENTIAL SAMAS HP-MU-P-1A--P2MM 2.21E-03 1.01 Makeup Pump (Standby) 1A in Maintenance Over half of the contribution from this event is related to seal LOCAs in which NSRW cooling to ICCW is lost. Providing an alternate means of cooling the ICCW heat exchangers would prevent the seal LOCAs in these sequences. FSW could be used as a backup cooling source for the ICCW heat exchangers. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action (SAMA 7). HL-V-7AV-7BCVAFD 2.03E-04 1.01 Line UP DHR HP Recrc MOV CCF Op V-7A;7B The low position of this event in the importance list indicates that hardware changes to specifically address the CCF of the DHR to HPI suction valves (DH-V-7A/B) would not be cost beneficial. The dominant contributor for this event is when it is paired with a small break LOCA alone (38% of contribution). In this case, the only options for mitigation appear to be the installation of a bypass line or an alternate DHR method.
A manually operated bypass would be effective assuming it was accessible, but a more appropriate approach for addressing this risk is believed to be through the seal LOCAs. Prevention of the seal/consequential LOCAs would preclude the need for HPR. The SAMAs suggesting the installation of high temperature, damage resistant seals (SAMA 2) and automated RCP trip logic (SAMA 8) would address the seal/consequential LOCAs contributors related to this event. OTHOT1_RCPTHP1OA 1.44E-02 1.01 OPERATOR FAILS TO TRIP REACTOR COOLANT PUMP ON LOSS OF NSCCW The contribution from the failure of this action could be reduced if high temperature sensors on the motor bearing cooling water lines were installed and used to provide automatic trip signals for the pumps (SAMA
 
8). RA-P-1AP-1BCP5FR 5.35E-05 1.01 DHRW Standby RW Pump CCF Run P5-
 
1A;1B The event is associated with loss of DHRW flow scenarios. Use of the NSCCW system to cool the DHR heat exchangers (DH-C-1A/B) would provide alternate heat removal capabilities (SAMA 3).
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-307 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS
%AC 4.48E-02 2.161 LOSS OF OFFSITE POWER Addressed by a similar event the Level 1 importance list. RECOFFSITEPWR 9.64E-01 1.698 OFFSITE POWER RECOVERED WITHIN 24 HOURS About 80% of the contributors including RECOFFSITEPWR are SBO events, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for RECOFFSITEPWR. An additional insight is that 80% of the contributors including RECOFFSITEPWR belong to RC8-01. These sequences are characterized by ex-vessel releases of corium and basemat failure. Ex-vessel release occurs due to lack of containment spray early while basemat failure largely occurs in spite of late recovery of containment sprays, which implies that early recovery of AC power would allow containment spray to prevent the ex-vessel release.
Environmental Report Appendix E  SAMA ANALYSIS Page E-308 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS MELT 5.00E-01 1.668 Likelihood That Water Pool in Cavity Will Not Stop Concrete Attack This event represents the probability that water will not prevent interaction between the core melt debris and the containment floor (containment has performed as designed, but the sprays cannot prevent containment damage). Over 50% of the cutset contributions including the event MELT are SBO events, which are represented by events RECOVERY-LOOP-03 and RECOVERY-
 
LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for MELT. An additional 20% to 25%
of the contributors are cases where the SBO EDG is available, but cannot be aligned in time to prevent a seal LOCA. These cases are addressed by SAMA 1. No potentially cost effective containment structure changes have been identified to address this issue (installation of a flooded rubble bed was estimated to be over $18 million for the ABWR [GE 1994]). RECSPRAYLT 9.99E-01 1.614 AVAILABILITY OF CONTAINMENT SPRAYS WITHOUT POWER DEPENDENCY RECSPRAYLT is completely tied to event RECOFFSITEPWR, which is addressed separately in this table. RECOVERY-LOOP-03 8.11E-02 1.355 NONRECOVERY OF OFFSITE POWER Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-309 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS RBSPRAY 9.99E-01 1.263 RB SPRAY SYSTEM IS AVAILABLE These cases are similar to RECSPRAYLT in that containment spray is ineffective at preventing containment failure. However, for these cases, AC power is available to support containment spray early. About 35% of the contributors including RECSPRAYLT also include the event MELT, which is addressed separately in this list. An additional 35% is related to containment over pressurization due to hydrogen burns. Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19). STREN1H2 5.00E-01 1.203 Likelihood That Cont Can Handle Comb. Gas Burn Press. W/ High Base Pressure This event represents the cases where a hydrogen burn occurs, but the containment does not fail due to the burn event. Over 99.5% of these cases include the event MELT. As for the
 
event MELT,  over 50% of the cutset contributions including the event MELT are SBO events, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for MELT. An additional 20% to 25% of the contributors are cases where the SBO EDG is available, but cannot be aligned in time to prevent a seal LOCA.
These cases are addressed by SAMA 1. CTMT-F-BENIGN 9.00E-01 1.17 CONTAINMENT LEAK BEFORE BREAK About 70% of the contributors including CTMT-F-BENIGN are related to hydrogen burns that fail containment. Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19). RECOVERY-LOOP-01 4.97E-01 1.158 NONRECOVERY OF OFFSITE POWER Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E  SAMA ANALYSIS Page E-310 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS RECOVERY-LOOP-04 4.97E-01 1.141 NONRECOVERY OF OFFSITE POWER Addressed by a similar event the Level 1 importance list. GB-EDG-1B---DGFR 2.07E-02 1.136 DIESEL 1B FAILS TO RUN Addressed by a similar event the Level 1 importance list.
DRYEFF 5.00E-01 1.127 Likelihood That Recombination Can Deplete Comb. Gas Given a Dry CavityThis event represents the cases where the hydrogen recombiners are able to remove enough hydrogen to prevent a catastrophic burn. As a result, early containment failure does not occur, but subsequent evolutions result in loss of containment integrity. Over 99.5% of these cases include the event MELT. As for the event MELT,  over 50% of the cutset contributions including the event MELT are SBO events, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for MELT. An additional 20% to 25% of the contributors are cases where the SBO EDG is available, but cannot be aligned in time to prevent a seal LOCA.
These cases are addressed by SAMA 1. GA-EDG-1A---DGFR 2.07E-02 1.127 DIESEL 1A FAILS TO RUN Addressed by a similar event the Level 1 importance list. NOSTREN1H2 5.00E-01 1.125 Likelihood That Cont Cannot Handle Comb. Gas Burn Press. W/ High Base Pressure These contributors are related to hydrogen burns that fail containment (for late containment failure). Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19). GA-1A1BSBO-CDGFR 1.53E-04 1.118 EDG CCF Run DG-1A;DG-1B;DG-SBO Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-311 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS EFEFP1------P7FR 5.06E-02 1.098 TURBINE-DRIVEN PUMP EF-P-1 FAILS TO RUN Addressed by a similar event the Level 1 importance list. NOEXSCRUBEFF 1.00E-01 1.096 Likelihood That Overlying Water Pool Will Not Scrub FPs Released From Corium This event is completely linked to the event MELT; however, the population of MELT events that it is associated wi th are not SBO events. About 30% are related to RECOVERY-LOOP-01 for which SAMAs 1 and 2 would be useful. The remaining contributors are a diverse mixture of LOCAs and transients that would not be mitigated by a single SAMA outside of the installation of an additional, independent DHR/injection system. Based on the high cost of a new DHR/injection system and the low contribution of all non-SBO transients and non-ISLOCAs to the MACR, this type of change would not be cost beneficial. No additional SAMAs are suggested to address this event. NODRYEFF 5.00E-01 1.09 Likelihood That Recombination Cannot Deplete Comb. Gas Given a Dry CavityThese contributors are related to hydrogen burns that fail containment. Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19). NOAFTSTREN1 5.00E-01 1.08 Likelihood That Cont Cannot Handle Comb. Gas Burn Press. W/ High Base Pressure These contributors are related to hydrogen burns that fail containment (for early containment failure). Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19). NOINERTAF 1.00E-01 1.08 Containment Has High Base Pressure Early After RV Failure Without Steam Inerting These contributors are related to hydrogen burns that fail containment (for early containment failure). Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19).
Environmental Report Appendix E  SAMA ANALYSIS Page E-312 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS
%LAIR 5.23E-03 1.08 LOSS OF AIR INITIATING EVENT Addressed by a similar event the Level 1 importance list. FLAG-SBOALIGN-1E 5.00E-01 1.074 SBO ALIGNED TO BUS 1E Addressed by a similar event the Level 1 importance list. JHHEF1-HBW1HEPOA 1.00E-04 1.073 EFHEF1_OPERH2HOA AND BWHBW1-----HP2OA Addressed by a similar event the Level 1 importance list. JHAHCD4RE27HEPOA 9.17E-05 1.07 AVHCD4_FF--HCDOA AND BWST-HRE27-HTKOA Addressed by a similar event the Level 1 importance list.
%TRIB 2.86E-03 1.069 INITIATING EVENT FOR SGTR ON OTSG B Addressed by a similar event the Level 1 importance list.
%TRIA 2.86E-03 1.069 INITIATING EVENT FOR SGTR ON OTSG A Addressed by a similar event the Level 1 importance list. GB-EG-Y-1B--DGMM 1.61E-02 1.069 Emergency Diesel Generator 1B in Maintenance Addressed by a similar event the Level 1 importance list.
GB1BDG------DGFS 1.13E-02 1.066 DIESEL GENE RATOR 1B FAILS TO START Addressed by a similar event the Level 1 importance list. GA-EG-Y-1A--DGMM 1.61E-02 1.065 Emergency Diesel Generator 1A in Maintenance Addressed by a similar event the Level 1 importance list.
GA1ADG------DGFS 1.13E-02 1.062 DIESEL GENE RATOR 1A FAILS TO START Addressed by a similar event the Level 1 importance list. WATEREFF 5.00E-01 1.061 Likelihood That Water in S/G Will Scrub Fission Products This event is related to SGTR scenarios. The failure to provide makeup to the BWST (BWST-HRE27-HTKOA) contributes to over 85% of the cutsets including WATEREFF. Event BWST-HRE27-HTKOA is addressed in the Level 1 importance list.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-313 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS NONCGASHIGH 1.00E-01 1.057 Likelihood That Non Condensable Gas Production is Not High Given a Dry Cavity This event is completely linked to the event MELT; however, the population of MELT events that it is associated with are not all SBO events. About 35% are related to RECOVERY-LOOP-03 and RECOVERY-LOOP-04, which are addressed by similar events in the Level 1 importance list.
Some additional benefit (about 25%) could be gained through the use of the RBEC system to provide alternate flow to the DHR heat exchangers (DH-C-1A/B) (SAMA 3). The remaining contributors are a diverse mixture of LOCAs and transients that would not be mitigated by a single SAMA outside of the installation of an additional, independent DHR/injection system, which is known not to be cost effective. No additional SAMAs are suggested to address this event. AV-LOCADV--HCDOA 1.00E+00 1.055 OPERATOR ACTION FAILURE TO LOCALLY OPERATE ADVS ON LOSS
 
OF AIR Addressed by a similar event the Level 1 importance list. NOWATEREFF 5.00E-01 1.055 Likelihood That Water in S/G Will Not Scrub Fission Products This event is related to SGTR scenarios. The failure to provide makeup to the BWST (BWST-HRE27-HTKOA) contributes to over 95% of the cutsets including WATEREFF. Event BWST-HRE27-HTKOA is addressed in the Level 1 importance list. GSHEO1A----HDGOA 2.66E-02 1.051 OPERATOR FAILS TO STARTSBODG Addressed by a similar event the Level 1 importance list. LOCA-SIZE-101 7.80E-01 1.05 PROBABILITY THAT RCP SEAL LOCA IS OF VSLOCA CATEGORY Addressed by a similar event the Level 1 importance list.
%VSB 2.56E-03 1.046 VERY SMALL BREAK LOCA Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E  SAMA ANALYSIS Page E-314 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS
%LNR 3.42E-03 1.044 LOSS OF NUCLEAR RIVER WATER Addressed by a similar event the Level 1 importance list. DABB1A------BYFD 4.84E-04 1.044 FAILURE OF BATTERY BANK 1A ON DEMAND Addressed by a similar event the Level 1 importance list. EF-CCFEFW-LETHAL 4.25E-04 1.043 LETHAL SHOCK TO THE EFW SYSTEM DUE TO COMMON CAUSE FAILURES Addressed by a similar event the Level 1 importance list. FLAG-SBOALIGN-1D 5.00E-01 1.041 SBO ALIGNED TO BUS 1D Addressed by a similar event the Level 1 importance list. NOHEATIML 1.00E-01 1.039 Prob. that Failure of the Primary System Does Not Occur Due to Heating Over 86% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for NOHEATIML. GS-SBODG----DGFR 2.07E-02 1.038 SBO DIESEL FAILS TO RUN Addressed by a similar event the Level 1 importance list. GAEDG-STARTCDGFS 5.25E-05 1.037 EDG Fail to Start CCF DG-All 3 Addressed by a similar event the Level 1 importance list. OP230KV-----OGFD 2.40E-03 1.034 LOSS OF 230KV TO AUX XFRMR 1A AND 1B Addressed by a similar event the Level 1 importance list. RARB-STANDBYFLAG 5.00E-01 1.034 BO TH DHRW TRAINS A AND B IN STANDBY Addressed by a similar event the Level 1 importance list.
%LGA 1.23E-03 1.032 LOSS OF GA POWER Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-315 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS INERTLT 1.00E-01 1.032 Sequence Late After RV Failure Has Low  Base Pressure From Gas Generation This event represents the cases where gas generation for the core melt process does not produce enough gas to create a high base pressure in the containment (related to evaluating consequences of a hydrogen burn). For the relevant cases (all RC8-01), the hydrogen burn does not cause containment failure, but subsequent evolutions result in loss of containment integrity. Over 99.8% of these cases include the event MELT. As for the event MELT,  over 60% of the cutset contributions including the event MELT are SBO events, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for MELT. An additional 28% of the contributors are cases where the SBO EDG is available, but cannot be aligned in time to prevent a seal LOCA. These cases are addressed by SAMA 1. INHINJ2_MUHHMUOA 1.00E+00 1.028 OPERATOR OPENS CROSS CONNECT VALVES MU-V-76A/B AND
 
STARTS MU-P-1C Addressed by a similar event the Level 1 importance list. NON-RECOV-LNR-IE 2.70E-01 1.025 NON-RECOVERABLE FRACTION OF
 
%LNR EVENTS Addressed by a similar event the Level 1 importance list. GADF-PALL6-CP2FS 3.62E-05 1.025 EDG Standby Pump CCF Start P2-ALL 6 Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E  SAMA ANALYSIS Page E-316 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS
%ISL 1.80E-07 1.024 INTERFACING SYSTEM LOCA For TMI-1, ISLOCA is dominated by DHR suction path failures after leak or rupture of valves DH-V-1 and DH-V-2. While the TMI-1 ISLOCA analysis does not take credit for any potentially mitigating actions, no actions that could reliably terminate the event are believed to be available. For example, 1) the isolation of DH-V-3 may not isolate the break or additional breaks may occur after isolation, 2) reduction of primary system pressure may reduce the flow out of the break, but it would not stop it, and 3) refill of the BWST does not place the plant in a stable state and the impacts of aux building flooding would have to be addressed. A potential SAMA would be to extend the high pressure boundary through valve DH-V-3 to allow an additional isolation point (SAMA 20). ISLOCA--COREMELT 1.00E+00 1.024 CORE DAMAGE DUE TO INTERFACING SYSTEM LOCA This event is completely tied to %ISL, which is treated separately on this list. GA-1A-1B---CDGFR 2.31E-04 1.023 EDG CCF Run DG-1A;DG-1B Addressed by a similar event the Level 1 importance list. GS-EG-Y-4---DGMM 1.30E-02 1.023 SBO Diesel Generator in Maintenance Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-317 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS CWNOLIMITLPME 1.00E-02 1.023 Plant Config and Layout Does Not Limit Material Reaching Cont. Wall With LPM The contributors including this event are composed of a diverse set of accident scenarios that lead to low pressure core melts. No single SAMA has been identified that would effectively eliminate a majority of the core damage sequences. Several SAMAs identified in the Level 1 importance list are applicable to portions of the contributors, but these issues are addressed by the Level 1 review and no new insights are available from the Level 2 cutsets for the core damage evolutions. The event CWLIMITLPME represents the probability that corium will not spread to the containment wall after a low pressure melt, which is described as "almost certain" in the L2 analysis based on the cavity configuration. The event here, CWNOLIMITLPME, is the complement of CWLIMITLPME. A possible plant enhancement would be to identify pathways that corium could reach the containment wall and to install shields to block the pathways or to flood the containment
 
early (SAMA 21). MF-MFPT----EVENT 2.09E-02 1.022 MFPT (LEGACY EVENT)
These events are related to the loss of MFW flow in after a trip when overcooling events have not
 
occurred. MFW and EFW availability are important to determining the status of fission product scrubbing for SGTR events and also for determining whether or not induced tube ruptures will occur. These events could be reduced in an independent AFW system were installed (SAMA 22). %SLT 4.22E-03 1.022 STEAM LINE BREAK IN TURBINE BUILDING Addressed by a similar event the Level 1 importance list.
Environmental Report Appendix E  SAMA ANALYSIS Page E-318 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS HP-_14A_14BCVAFD 2.03E-04 1.021 HPI Train Fails MOV CCF Op MU-V-14A;14B Addressed by a similar event the Level 1 importance list.
%LNS 2.74E-03 1.021 LOSS OF NUCLEAR SERVICES CLOSED COOLING WATER Addressed by a similar event the Level 1 importance list.
%FW 5.40E-02 1.02 LOSS OF FEEDWATER These events are related to the loss of MFW flow in after a trip followed by failure of EFW and induced SGTR. MFW and EFW availability are important to determining the status of fission product scrubbing for SGTR events and also for determining whether or not induced tube ruptures will occur. These events could be reduced in an independent AFW system were installed (SAMA 22). GSEG-Y-4----DGFS 1.13E-02 1.02 ST ATION BLACKOUT DG FAILS TO START Over 99% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for GSEG-Y-4----DGFS. CFRR-V-6----VCFF 1.62E-02 1.018 RR-V6 FAILS TO OPERATE This valve failure is related to the loss of RBEC return flow for containment cooling. The TMI-1 HRA documentation indicates that there are no alarm response procedures related to low flow on the system that woulddirect the operators to open the bypass valve (RR-V-5) when RR-V-6 fails to open. A potential SAMA would be to develop procedures to direct oper ation of the bypass valve when the normal return path fails (SAMA 23).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-319 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS CFHRR1-----HVAOA 7.79E-01 1.018 OPERATOR FAILS TO OPEN MOV RR-V-5 This valve failure is related to the loss of RBEC return flow for containment cooling. The TMI-1 HRA documentation indicates that there are no alarm response procedures related to low flow on the system that woulddirect the operators to open the bypass valve (RR-V-5) when RR-V-6 fails to open. A potential SAMA would be to develop procedures to direct oper ation of the bypass valve when the normal return path fails (SAMA 23). RECOVERY--LNR-IE 7.30E-01 1.018 RECOVERABLE FRACTION OF
%LNR EVENTS Addressed by a similar event the Level 1 importance list. JHHRE27HL1AHEPOA 2.00E-04 1.017 BWST-HRE27-HTKOA AND DLHHL1A----HVHOA Automating BWST refill would effectively eliminate this JHEP and provide a reliable means of maintaining level in the BWST (SAMA 10). NORECOFFSITEPWR 3.60E-02 1.017 OFFSITE POWER NOT RECOVERED WITHIN 24 HOURS Most of the contributors including this event result in late containment failure due to over pressurization. Over 70% of the contributors are SBO cases, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for NORECOFFSITEPWR. BWHBW1-----HP2OA 2.18E-03 1.017 OPERATOR FAILS TO INITIATE HPI Addressed in the Level 1 importance list through dependent operator action terms JHHEF1-HBW1HEPOA and JHHAMHEFHBWHEPOA. JHHOT1-XTIEHEPOA 5.10E-02 1.017 OTHOT1_RCPTHP1OA AND NR-NRSRXTIEHVAOA Automating RCP trip on high cooling water temperature would effectively eliminate this JHEP and provide a reliable means of preventing pump/seal damage (SAMA 8).
Environmental Report Appendix E  SAMA ANALYSIS Page E-320 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS GEOMFREEZE 5.00E-01 1.016 Cavity Geometry Allows Enough Corium to Disperse For Freezing Over 87% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-
 
LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for GEOMFREEZE. No potentially cost effective containment structure changes to impact the dispersal of corium in the cavity have been identified. JHHHL1AHSR2HEPOA 2.00E-04 1.016 DLHHL1A----HVHOA AND SAHSR2-----HSROA Addressed by a similar event the Level 1 importance list. BWST-HRE27-HTKOA 2.65E-02 1.015 FAILURE TO REFILL BWST (SPLIT FRAC REV) Addressed in the Level 1 importance list through dependent operator action JHAHCD4RE27HEPOA. INMU-P-1C--HMUOA 1.00E+00 1.014 OPERATOR FAILURE TO ALIGN AND START MU-P-1C Addressed by a similar event the Level 1 importance list. CTMT-F-NOTBENIGN 1.00E-01 1.014 PR OBABILITY THAT CONTAINMENT FAILURE IS NOT BENIGN This event represents the probability that containment failure due to over pressurization will be a failure that results in a rapid blowdown of containment. Over 80% of the contributors including CTMT-F-NOTBENIGN are failure due to hydrogen burns. Installation of battery backed hydrogen igniters would reduce the contribution from these events (SAMA 19).
%SBL 4.50E-04 1.014 SMALL BREAK LOCA Addressed by a similar event the Level 1 importance list. JHHAM2-HEF1HEPOA 4.61E-03 1.014 AMHAM2-----HC1OA AND EFHEF1_OPERH2HOA This dependent operator action term is addressed by SAMA 13, which would automate operator action AMHAM2-----HC1OA and preclude the need for EFHEF1_OPERH2HOA. No additional SAMAs are required.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-321 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS HADC-V-2A---VCFT 3.00E-03 1.013 DC-V2A FAILS TO REMAIN OPEN Addressed by a similar event the Level 1 importance list. HADC-V-65A--VCFT 3.00E-03 1.013 DC-V65A TRANSFERS TO  DIFFERENT STATE Addressed by a similar event the Level 1 importance list. HP-MU-P-1B--P2MM 7.46E-03 1.012 Makeup Pump (Operating) 1B in Maintenance Addressed by a similar event the Level 1 importance list. DXBATT1A1B-CBYFF 3.51E-06 1.012 Batteries 1A and 1B CCF Operate The importance of this event is driven by its contribution to containment isolation failure in SBO cases (dominated by RECOVERY-LOOP-04), which is dependent on AC power. These sequences could be mitigated by preventing core damage in the same manner as suggested for RECOVERY-LOOP-04. DX-1-ABCD--CBCFF 3.39E-06 1.012 Battery Charger CCF of 3/4 and 4/4 The importance of this event is driven by its contribution to containment isolation failure in SBO cases (dominated by RECOVERY-LOOP-04), which is dependent on AC power. These sequence could be mitigated by preventing core damage in the same manner as suggested for RECOVERY-LOOP-04. JHHOTHMRXTIHEPOA 3.10E-03 1.011 OTHOT1_RCPTHP1OA; MRHMR1-----HMUOA; NR-NRSRXTIEHVAOA Addressed by a similar event the Level 1 importance list.
%RT 4.82E-01 1.011 REACTOR TRIP The importance of this event is driven by a diverse set of contributors that are addressed elsewhere in the importance lists, including OP230KV-----OGFD, MELT, RECOFFSITEPWR, and RECSPRAYLT. No single, potentially cost beneficial S AMA has been identified to mitigate all of the risk associated with the "reactor trip" initiating event.
Environmental Report Appendix E  SAMA ANALYSIS Page E-322 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS HA-P-1AP-1BCP2FS 1.50E-04 1.011 DH Clsd Cool Stdby Pmp CCF Strt P2-1A;1B Addressed by a similar event the Level 1 importance list. NRHNS8A----HP1OA 5.37E-01 1.011 OPERATOR FAILS TO ISOLATE FAILED RW PUMP (POWER UNAVAILABLE) Addressed by a similar event the Level 1 importance list. GADF-PALL6-CP2FR 1.60E-05 1.011 EDG Standby Pump CCF Run P2-ALL6 Over 99.5% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for GADF-PALL6-CP2FR. OP-OPB-CONDITION 3.00E-01 1.011 POWER SUPPLY UNAVAILABLE GIVEN A TURBINE BYPASS SIGNAL Addressed by a similar event the Level 1 importance list. NRNR-V-20A--VPFD 1.35E-03 1.011 CHECK VALVE NR-V20A FAILS TO RESEAT Addressed by a similar event the Level 1 importance list. HA-DC-P-1A--P1MM 2.84E-03 1.011 Decay Heat Closed Cycle Cooling Water Pump 1A in Maintenance Addressed by a similar event the Level 1 importance list. GSFS-V-646--VCFD 6.38E-03 1.01 AIR OPERATED VALVE FS-V-646
 
FAILS ON DEMAND This event causes the failure of the cooling flow to the SBO EDG and over 99.5% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed by similar events in the Level 1 importance list and the same SAMAs are applicable for GSFS-V-646--VCFD.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-323 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS GSFS-V-647--VCFD 6.38E-03 1.01 AIR OPERATED CONTROL VALVE FS-V-647 FAILS ON DEMAND This event causes the failure of the cooling flow to the SBO EDG and over 99.5% of the contributors including this event are SBO scenarios, which are represented by events RECOVERY-LOOP-03 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for GSFS-V-647--
VCFD. HADC-P-1A---P2FS 2.46E-03 1.01 DHCCW PUMP DC-P1A FAILS TO START Addressed by a similar event the Level 1 importance list. HBDC-V-2B---VCFT 3.00E-03 1.01 DC-V2B FAILS TO REMAIN OPEN Addressed by a similar event the Level 1 importance list. HBDC-V-65B--VCFT 3.00E-03 1.01 DC-V65B TRANSFERS TO  DIFFERENT STATE Addressed by a similar event the Level 1 importance list. JHHAMHEFHBWHEPOA 2.40E-04 1.01 JHHAM2-HEF1HEPOA AND BWHBW1-----HP2OA Addressed by a similar event the Level 1 importance list.
SPARKAFT_1 1.00E-01 1.01 PROB THAT SPARK IS AVAILABLE EARLY AFTER RV FAILURE WITHOUT RB SPRAY These cases are related to evolutions in which an ignition source is available and causes a non-catastrophic hydrogen burn. Containment failure occurs later due primarily to basemat failures. For the contributors including this event, most of the contribution results from core damage events that could have been mitigated if it were possible to swap the train to which the SBO EDG was aligned after equipment failure. This is addressed by SAMA 1.
Environmental Report Appendix E  SAMA ANALYSIS Page E-324 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-2 LEVEL 2 IMPORTANCE LIST REVIEW EVENT NAME PROBABILITYRED W DESCRIPTION POTENTIAL SAMAS DABATTCHGR-HBCOA 1.00E-01 1.01 HEP FOR FAILURE TO ALIGN SPARE CHARGER 1E OR 1F About 80% of the contributors including DABATTCHGR-HBCOA are LOOP events that include events RECOVERY-LOOP-01 and RECOVERY-LOOP-04. These events are addressed in the Level 1 importance list and the same SAMAs are applicable for DABATTCHGR-HBCOA. CWNOLIMITHPME 1.00E-01 1.01 Plant Config and Layout Does Not Limit Material Reaching Cont. Wall
 
With HPM About 70% of the contribution from this event is linked to "AV-LOCADV--HCDOA", which is addressed by a similar event the Level 1 importance list. As discussed there, if existing procedures are credited, the contribution from AV-LOCADV--HCDOA will be greatly reduced, which implies that event CWNOLIMITHPME would not remain above the RRW review threshold of 1.01. However, SAMA 21 was developed for a similar event (CWNOLIMITLPME) and it addresses the same issues relevant to CWNOLIMITHPME. HB-DC-P-1B--P1MM 2.84E-03 1.01 Decay Heat Closed Cycle Cooling Water Pump 1B in Maintenance Addressed by a similar event the Level 1 importance list. EF-EF-P-1---P1MM 6.57E-03 1.01 EFW Pump (Turbine Driven) 1 in Maintenance About 90% of these events are SBO cases, represented by RECOVERY-LOOP-04. This event is addressed in the Level 1 importance list and the same SAMAs are applicable for EF-EF-P-1---P1MM.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-325 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 1 Enhance the SBO EDG for Auto Alignment and Loading The current capability of the SBO EDG is limited by manual actions to diagnose and respond to conditions requiring a start of the SBO EDG. While the time required to start and load the EDG is relatively short, it is close enough to the 13 minute limit for restoration of seal cooling after a total loss that no credit is taken for the SBO EDG to prevent seal LOCAs in LOOP evolutions with normal EDG failures. Automation of SBO EDG operation would reduce the time required to restore seal cooling and through this function, a large portion of the seal LOCA CDF could be eliminated. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $3,125,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.1).
Environmental Report Appendix E  SAMA ANALYSIS Page E-326 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 2 Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for Extended EFW Operation Currently, alternate RCP pump seals are available that can effectively prevent seal LOCAs caused by loss of RCP seal cooling (Flowserve N-9000 seals). It is estimated that these seals
 
will limit leakage flow to about 1 gpm per seal on loss of cooling, which is low enough to maintain core coverage in cases where seal LOCAs would normally result in core uncovery/core damage within the PRA's 24 hour mission time. The ability to prevent a seal LOCA will allow for extended operation in SBO conditions if level instrumentation can be supplied using the vital 120V AC system. Powering the station battery chargers with a portable 480V AC generator would provide this capability and allow control of the TD EFW system to be retained in the MCR. Level 1 TMI-1 Importance
 
List The cost of this enhancement was estimated to be $7,300,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.2).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-327 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 3 Use NSCCW as an Alternate Cooling Source for the DHR Heat Exchangers (DH-C-1A/B) For LOCAs requiring heat removal with the RHR system, DHRW and DHCCW failures are large contributors to loss of the primary cooling function. Providing the ability to cross-tie the NSCCW system to the DHR heat exchangers would diversify the plant's heat removal capability and eliminate the failures associated with loss of DHRW or DHCCW flow. The hard piped connections are assumed to be sized to allow enough flow to remove decay heat (not just pump cooling loads) and that each division is provided with a cross-connection. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $2,450,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.3).
Environmental Report Appendix E  SAMA ANALYSIS Page E-328 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 4 Provide Alternate Power to HPI Pump Minimum Flow Recirculation Valves MU-V-36 and MU-V-
 
37 The current PRA model logic correctly assumes isolation of valves MU-V-36 and 37 on an ESAS, but it does not include the AC power dependences for the "close" action. However, the logic related to opening the minimum flow valves does include the power dependences, which can result in the generation of cutsets that include the failure to open a flow path that was never isolated. If the appropriate power dependencies were accounted for in the isolation logic, the only events that could cause the MU-V-36 or MU-V-37 valves to be "stranded closed" are those in which an ESAS occurs when both divisions of power are available and then division "A" power fails before MU-V-36 can be opened. Level 1 TMI-1 Importance
 
List Not Required (screened on PRA
 
insights). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis(refer to Section E.6.4
).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-329 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 5 Enhance Valves MU-V-76A/B and MU-V-77A/B to
 
Allow for Rapid Alignment Changes in Accident
 
Conditions The current MU-V-76A/B and MU-V-77A/B valve configurations do not allow for rapid re-alignment during accident conditions. For TMI-1, the capability to quickly align the "C" HPI pump for seal injection would reduce the risk of prominent accident sequences in which thermal barrier cooling has failed in conjunction with the "A" and "B" HPI pumps. Replacing MU-V-76A/B and MU-V-77A/B with MOVs operable from the main control room would allow TMI-1 to use the "C" HPI pump for seal injection and prevent seal LOCAs when the normal cooling methods are unavailable. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $3,150,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.5). 6 Add Cross-ties Within the Trains of the Cooling Systems -DHR
-DHRW -DHCCW Some failure combinations that eliminate both trains of the DHR related cooling systems could be mitigated if cross-ties were available between trains of the DHR, DHRW, and DHCCW systems (not between the systems). For example, these cross-ties would be helpful in conditions where the flow path fails in one train while a pump failure or maintenance event disables the opposite train. To ensure the DHR cross-ties can be implemented in a timely manner for LPI requirements, the associated valves should be operable from the main control room. Level 1 TMI-1 Importance List The cost of installing the powered DHR cross-tie was estimated to be $2,750,000 by the TMI staff (Exelon 2007c). The cross-ties for the DHCCW and DHRW systems are not required to be MOVs due to the longer times available for performing the cross-tie and while there would be a substantial additional cost related to the addition of these cross-ties, only the DHR cross-tie cost of $2,750,000 is used here based on
 
the availability of information. Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.6).
Environmental Report Appendix E  SAMA ANALYSIS Page E-330 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 7 Use Fire Service Water as an Alternate Cooling Source for the ICCW Heat Exchangers For cases in which NSRW is unavailable due to hardware failures (e.g., flow diversion), the Fire Service Water system could be used to directly cool the ICCW heat exchangers for thermal barrier cooling support. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action. Level 1 TMI-1 Importance
 
List Palisades estimated $2.9 million for Fire water cooling to CCW HXs (NMC 2005), Calvert Cliffs estimated $565k for alt DHR cooling (BGE 1998), and Brown's Ferry estimated $1 million for Fire Water to DHR HXs (TVA 2003). The Brown's Ferry estimate is used for TMI. Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.7). 8 Automate Reactor Coolant Pump Trip Seal LOCAs resulting from operator failures to trip the RCPs on loss of motor bearing cooling could be reduced if high temperature sensors were installed on motor bearing cooling water lines to provide automatic trip signals. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $145,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.8). 9 Proceduralize Local ADV Operation TMI-1 has procedures to perform the local ADV operations that are not credited in the PRA model (the failure probability is set to 1.0). If the available procedures are credited, the RRW value of the operator action would be reduced below the SAMA review threshold. This SAMA is used demonstrate the reduction in the RRW that would occur when a reasonable failure probability is applied to the operator action. Level 1 TMI-1 Importance List Not Required (screened on PRA insights). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.9).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-331 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 10 Automate BWST Refill Failure to refill the BWST is a large contributor to some SGTR sequences, especially those in which the MS ADVs fail to operate. Automating the refill function would improve the reliability of this process and reduce the contributions from prominent SGTR sequences by providing a long term high pressure injection source. This SAMA requires a new pump with a flow rate of at least 400 gpm with a connection to a borated water source that will provide suction for 24 hours. In addition, the pump should be able to supply water from a non-borated water source for an indefinite periods of time after depletion of the borated water source. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $3,800,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.10). 11 Enhance Extreme External Flooding
 
Mitigation Equipment to Address SBO and Loss of Seal Cooling Scenarios Making the extreme flooding equipment useful for SBO conditions, especially those with TD EFW failure, would require permanently mounting the submersible pumps so that the suctions could easily be swapped from a piped water source to the flood water source. Permanently installing the portable generator and the pumps so that they could be aligned from the MCR would improve alignment
 
capabilities and add ress non-SBO loss of seal cooling cases through the
 
ability to rapidly align alternate seal cooling. Level 1 TMI-1 Importance
 
List The cost of this enhancement was estimated to be $4,250,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.11).
Environmental Report Appendix E  SAMA ANALYSIS Page E-332 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 12 Use the DHR System as an Alternate Suction Source for HPI Failures of the BWST suction path to the HPI pumps will lead to core damage in scenarios requiring early makeup. Through implementation of procedure changes, the DHR system could be aligned to take suction from the BWST and supply flow to the HPI system to allow injection in these cases. Level 1 TMI-1 Importance
 
List This change can be implemented at TMI-1 through only procedure changes as no interlocks are associated with the suggested alignment. Procedure changes are estimated to cost about $50,000 (CPL 2004). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.12). 13 Change IA System Logic to Automatically Start IA-P-1A/B After a Low Voltage Trip in Conjunction with an ESAS The current IA system logic requires the operators to re-load the IA compressors on emergency power after a low voltage trip when an ESAS is registered. Automating the re-loading of these compressors would remove the requirement for the operators to perform this task in accident conditions. Level 1 TMI-1 Importance
 
List The cost of this enhancement was estimated to be $950,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.13).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-333 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 14 Replace HPI Pump Cooling Alignment Valves with MOVs In the event that the normally aligned cooling source to a HPI pump fails, the current plant configuration requires local operation of the valves to swap the pump to the alternate cooling source. The time required to perform this action is considered to preclude it as a means of both preventing seal LOCAs in loss of seal cooling evolutions and for providing high pressure makeup. Replacing the valves with MOVs would allow the operators to rapidly align the alternate cooling source from the MCR in time to prevent a seal LOCA or provide high pressure injection. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $3,150,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.14).
Environmental Report Appendix E  SAMA ANALYSIS Page E-334 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 15 Automate Swap to Recirculation Mode The operator action to swap to recirculation mode is a key action for LOCA scenarios. Automating this function would improve the reliability of this action, especially in the rapidly evolving events where other actions are competing for the attention of the operators. Level 1 TMI-1 Importance
 
List Multiple SAMA analyses have included estimates for this type of change, but the estimates vary by over a factor of 3.5: - Oconee estimated the cost at
 
over $1 million per unit (DUKE 1998)) - Point Beach estimated the cost at over $1 million per unit (NMC 2004) - Catawba estimated the cost at
 
over $1 million (DUKE 2001) - Turkey Point estimated the cost to be about $450,000 (per unit)
(FPL 2000) - H.B. Robinson $265,000 (single unit) (CPL 2002) For TMI-1, the $450,000 estimate from Turkey Point is used as it is in the middle range of the industry estimates identified. Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.15). 16 Automate HPI Injection on Low Pressurizer Level Providing an automatic signal to initiate HPI on low pressurizer level would improve the reliability of HPI initiation. Level 1 TMI-1 Importance List The cost of this enhancement was estimated to be $1,100,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.16).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-335 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 17 Auto Isolate Steam Generators on High Steam Line Flow For steam line breaks downstream of the MSIVs, failure to isolate the relevant steam generator is an important contributor to core damage. The addition of logic to isolate the steam generator on high steam line flow would reduce the core damage contribution from isolation failures. Level 1 TMI-1 Importance List This SAMA is considered to be similar in scope to SAMA 13 and the same cost of implementation
($950,000) is used for this SAMA. Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.17). 18 Provide the Capability to Align the Standby Battery Charger and the 1A/1B Cross-tie from the MCR TMI has a spare 125V DC battery charger for each division that can be aligned to either battery bank within a division in the event that a normally operating battery charger fails. Currently, the alignment requires local actions. There is typically adequate time to align the charger in the event of a failure, but additional changes could be made to allow rapid alignment of the spare charger from the MCR to reduce the manipulation time and improve the man-machine interface. Level 1 TMI-1 Importance List No plant specific implementation cost was developed for this SAMA. Based on the low impact of the SAMA, the $100,000 minimum cost of a hardware modification (Exelon 2003) is used as the implementation cost. Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.18). 19 Install Battery Backed Hydrogen Igniters or a Passive Hydrogen Ignition System The addition of igniters would provide a means of preventing catastrophic combustible gas burns by continuously burning these gases before they reach critical levels. Providing battery backup power would increase the likelihood that this system would be available in LOOP events. Use of a passive system would also function in LOOP as well as long term SBO scenarios. Level 2 TMI-1 Importance
 
List The cost of this enhancement was estimated to be $760,000 in the Calvert Cliffs SAMA analysis (BGE 1998). Cannot be screened on cost or applicability to theplant. Retain for Phase II analysis (ref e r to Sect i on E.6.19).
Environmental Report Appendix E  SAMA ANALYSIS Page E-336 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 20 Extend the High Pressure Boundary Through DHR Valve DH-V-3 for ISLOCA Isolation The highest frequency ISLOCA scenario for TMI-1 is through two
 
valves in the DHR suction line. While the scenario's CDF is low, the release frequency is relatively high given that primary containment is bypassed by definition. No effective mitigating actions are considered to be available in these cases because 1) the break may occur upstream of DH-V-3 or additional breaks in the low pressure boundary may occur after closure of a low pressure isolation valve, 2) reduction of primary system pressure may reduce the flow out of the break, but it would not stop it, and 3) refill of the BWST does not place the plant in a stable state and results in auxiliary building flooding. Extending the pressure boundary through DH-V-3 would provide an additional isolation point in these cases. Level 2 TMI-1 Importance
 
List The cost of this enhancement was estimated to be $3.030,000 by the TMI staff (Exelon 2007c).
Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.20).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-337 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 21 Install Concrete Shields to Block Direct Pathways from the RPV to the Containment Wall and/or Direct Containment Flooding Early in External Flooding Scenarios This SAMA is based on a failure mode identified in the Level 2 analysis that indicates corium ejection during RV failure could result in dispersal of debris such that it could directly interact with the containment wall and cause a failure of the wall. For some external flooding scenarios, it may be possible to change the procedures to direct containment flooding early such that water would be available on the containment floor before loss of power.Level 2 TMI-1 Importance List The cost of this enhancement was estimated to be $1,200,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.21). 22 Install an Independent AFW System For TMI-1, loss of MFW after a trip coupled with loss of EFW can lead to large radionuclide releases in SGTR and induced SGTR scenarios due to the unavailability of water in the SGs for fission product scrubbing. A large contributor to EFW failure is estimated to be system wide common cause failures. An independent, motor driven, auxiliary feedwater system would be an effective means of addressing these cases. Power dependence is not a large issue for the cases addressed by this SAMA and the independent EFW pump is assumed to be powered by existing emergency power such that it would not be capable of mitigating SBO scenarios. Level 2 TMI-1 Importance List Calvert Cliffs estimated the cost of installing an additional HPSI pump with a dedicated diesel to be between $5 million and $10 million (BGE 1998). This type of enhancement is similar is scope to the changes required for this SAMA and the lower bound
 
estimate of $5 million is used for this SAMA as the diesel generator is not required for this SAMA. Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.22).
Environmental Report Appendix E  SAMA ANALYSIS Page E-338 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 23 Develop Alarm Response Procedures to Direct Operation of RR-V-5 on Low RBEC Flow  Failure of RR-V-6 to open results in the loss of RBEC flow to the reactor building coolers, which can be diagnosed using the system flow indicators in the main control room; however, no alarm response procedures exist to specifically direct operation of the bypass valve (RR-V-5). If this procedure was developed, it may reduce the diagnosis time and improve the reliability of this operator action in an accident conditions. Level 2 TMI-1 Importance
 
List Procedure changes are estimated to be $50,000 (CPL 2004). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.23).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-339 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 24 Install Damage Resistant, High Temperature RCP Seals with a Diesel Engine as an Alternate Drive for an EFW Pump and a Portable Generator for Level
 
Control Instrumentation For SBOs in which EFW has failed, neither primary nor secondary side cooling is available. Installing the enhanced RCP seals will prevent seal LOCAs and use of a portable generator would allow the turbine driven EFW pump to be used for extended periods in an SBO, as suggested in SAMA 2. However, in the event that the turbine driven EFW pump fails, there would be no means of providing secondary side makeup.
Turbine driven EFW failures could be mitigated if an engine was available to drive one of the EFW pumps. Other industry SAMA applications have suggested similar strategies, but they typically suggest the turbine driven pumps as the best option for connection to the engine based on ease of connection. For scenarios with turbine driven EFW failure, however, the initial TD EFW pump failure may prevent its further use even with an alternate motive source. As a result, this SAMA, in addition to the requirements of SAMA 2, requires that the diesel engine be connected to one of the motor driven EFW pumps. Palisades SAMA Analysis (NMC 2005) The cost of implementation for this SAMA is estimated to be a combination of SAMA 2
($7,300,000) and the $1.1 million estimate for a direct drive diesel injection pump from Palisades (NMC 2005). The total implementation cost is $8,400,000. Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.24).
Environmental Report Appendix E  SAMA ANALYSIS Page E-340 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 25 Install an Additional EDG An additional source of AC power is a potential means of supplying an entire division of safety equipment in the event that on-site AC power is lost in a LOOP. While additional EDGs are expensive, they can be cost effective at some plants, especially those with a large LOOP/SBO contribution to CDF. Palisades
 
SAMA Analysis (NMC 2005) Brown's Ferry estimated the cost of installing an additional EDG to be $6 million (TVA 2003). While there are estimates as high as $25 million used in SAMA analyses for the installation of additional EDGs, the Browns Ferry estimate is used for TMI-1. Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.25). 26 Reroute Cables so that They Do Not Pass Over Ignition Sources in Fire Area CB-FA-2e (West Inverter Room) or Wrap them in Fire Proof Material Some of the risk from fires in this room is from damage to cables that run over ignition sources. If the cable trays were re-routed away from the electrical equipment that they currently pass over, the consequences of equipment fires in the inverter room could be reduced. TMI-1 IPEEE (Fire) Of the two options, cable wrapping was determined to be the more cost effective approach.
The cost of performing the cable wrapping in CB-FA-2e was estimated to be $900,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.26). 27 Improve the 480V AC load center welds The IPEEE determined that the existing 480V AC load centers were among the weaker components in the TMI-1 AC distribution system. Adding reinforcements to the welds on the load center framework would improve the seismic durabilit y of the structure and increase the likelihood that the system would be available after a seismic event. The other low seismic capacity components, the EDG air receivers, were enhanced subsequent to the completion of the IPEEE.
TMI-1 IPEEE (Seismic) The cost of this enhancement was estimated to be $575,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.27).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-341 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 28 Improve the Decay Heat Service Cooler (DC-C-2A/B)
Anchorages The IPEEE determined that the existing Decay Heat Service Coolers (DC-C-2A/B) lacked sufficiently durable anchorages. Replacing the anchorages with more robust anchorages would improve the seismic
 
durability of the stru cture and increase the likelihood that the heat exchangers would be available after a seismic
 
event. TMI-1 IPEEE (Seismic) The cost of this enhancement was estimated to be $575,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.28). 29 Replace EDG Ground Resistors Failure of the EDG ground resistors results in failure of the EDGs, which will lead to core damage in the event that off-site power is not available. Given that the HCLPF capacity for these components was estimated at 0.25g compared with 0.09g capacities of off-site power components (such as the 1/A and 1/B distribution buses or the aux transformers), it is likely that core damage will ensue due to long term loss of power if the EDG ground resistors fail from seismic shock.
Replacing the resistors with more durable versions would improve the reliability of the EDGs in seismic
 
events. TMI-1 IPEEE (Seismic) The cost of this enhancement was estimated to be $800,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.29).
Environmental Report Appendix E  SAMA ANALYSIS Page E-342 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 30 Improve Diesel Fire Pump Fuel Oil Tank and Battery Rack Supports The Fire Service Water system provides cooling to the SBO EDG, backup cooling the DHCCW heat exchangers, and backup cooling to the "1A" and "1B" Instrument Air compressors. While seismic failures to the systems FSW supports would likely limit the benefit of improving the fuel oil tank and battery racks, some benefit may be available through improvements to the diesel fire pump's
 
reliability.
TMI-1 IPEEE (Fire/Seismic) The cost of this enhancement was estimated to be $150,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.30).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-343 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 31 Modify Specific Containment Penetration MOVs to "Fail Closed" Most containment penetrations have AOV or SOV isolation valves that will fail closed on loss of air or power; however, there are cases in which MOVs are used instead. Those lines that do not include a pair of AOVs or
 
SOVs that fail closed are typically below 1" in diameter or include at least one AOV or SOV that will fail closed on loss of air or power. However, the NSCCW and RBEC systems include penetrations that only include MOVs.
While these are closed cooling systems that would not normally provide a credible release path, heat exchanger breaks in seismic events could provide containment bypass routes in the event that a failure also occurs in the reactor building. Changing one of the valves in each of these paths to fail closed is a means of increasing the isolation probability over what is available from manual action.
TMI-1 IPEEE (Seismic) The cost of this enhancement was estimated to be $4,100,000 by the TMI staff (Exelon 2007c). Cannot be screened on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.31).
Environmental Report Appendix E  SAMA ANALYSIS Page E-344 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-3 PHASE I SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE COST ESTIMATE PHASE I DISPOSITION 32 Pre-stage Severe Flooding Equipment Pre-staging the equipment used to prevent core damage in severe flooding conditions would reduce sources of error in the alignment actions and reduce the time required to perform the task. Potential changes include: - Storing the portable EDG on the turbine deck - Adding a normally empty fuel oil tank for the portable EDG to the turbine
 
deck - Permanently running power cable from the portable EDG to the pump areas  A potential permutation of this SAMA would be to procure an additional portable EDG to reduce the failure contribution from the power source.
TMI-1 IPEEE (External Flooding) The cost of implementation is estimated to be $1,700,000 (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.32). 33 Increase the Flood Protection Height The current configuration protects to the design basis limit of 310 feet msl and levels any higher result in topping of the existing flood doors and flooding of sensitive areas. Raising the height of the flood doors (or completely sealing the doors) would prevent water incursion and allow for continued
 
operation of the normal safety equipment.
TMI-1 IPEEE (External Flooding) The cost of this enhancement was estimated to be $2,700,000 by the TMI staff (Exelon 2007c). Cannot be screened
 
on cost or applicability to the plant. Retain for Phase II analysis (ref e r to Sect i on E.6.33).
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-345 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 1 Enhance the SBO EDG for Auto Alignment and Loading The current capability of the SBO EDG is limited by manual actions to diagnose and respond to conditions requiring a start of the SBO EDG. While the time required to start and load the EDG is relatively short, it is close enough to the 13 minute limit for restoration of seal cooling after a total loss that no credit is taken for the SBO EDG to prevent seal LOCAs in LOOP evolutions with normal EDG failures. Automation of SBO EDG operation would reduce the time required to restore seal cooling and through this function, a large portion of the seal LOCA CDF could be eliminated. Level 1 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial". 2 Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for
 
Extended EFW Operation Currently, alternate RCP pump seals are available that can effectively prevent seal LOCAs caused by loss of RCP seal cooling (Flowserve N-9000 seals). It is estimated that these seals will limit leakage flow to about 1 gpm per seal on loss of cooling, which is low enough to maintain core coverage in cases where seal LOCAs would normally result in core uncovery/core damage within the PRA's 24 hour mission time. The ability to prevent a seal LOCA will allow for extended operation in SBO conditions if level instrumentation can be supplied using the vital 120V AC system. Powering the station battery chargers with a portable 480V AC generator would provide this capability and allow control of the TD EFW system to be retained in the MCR. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E  SAMA ANALYSIS Page E-346 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 3 Use NSCCW as an Alternate Cooling Source for the DHR Heat Exchangers (DH-C-1A/B) For LOCAs requiring heat removal with the RHR system, DHRW and DHCCW failures are large contributors to loss of the primary cooling function. Pr oviding the ability to cross-tie the NSCCW system to the DHR heat exchangers would diversify the plant's heat removal capability and eliminate the failures associated with loss of DHRW or DHCCW flow. The hard piped connections are assumed to be sized to allow enough flow to remove decay heat (not just pump cooling loads) and that each division is provided with a cross-connection. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 4 Provide Alternate Power to HPI Pump Minimum Flow Recirculation Valves MU-V-36 and MU-V-37 The current PRA model logic correctly assumes isolation of valves MU-V-36 and 37 on an ESAS, but it does not include the AC power dependences for the "close" action. However, the logic related to opening the minimum flow valves does include the power dependences, which can result in the generation of cutsets that include the failure to open a flow path that was never isolated. If the appropriate power dependencies were accounted for in the isolation logic, the only events that could cause the MU-V-36 or MU-V-37 valves to be "stranded closed" are those in which an ESAS occurs when both divisions of power are available and then division "A" power fails before MU-V-36 can be opened. Level 1 TMI-1 Importance List Screened from analysis based on PRA insights as described in Section E.6.
: 4. 5 Enhance Valves MU-V-76A/B and MU-V-77A/B to Allow for Rapid Alignment Changes in Accident Conditions The current MU-V-76A/B and MU-V-77A/B valve configurations do not allow for rapid re-alignment during accident conditions. For TMI-1, the capability to quickly align the "C" HPI pump for seal injection would reduce the risk of prominent accident sequences in which thermal barrier cooling has failed in conjunction with the "A" and "B" HPI pumps. Replacing MU-V-76A/B and MU-V-77A/B with MOVs operable from the main control room would allow TMI-1 to use the "C" HPI pump for seal injection and prevent seal LOCAs when the normal cooling methods are unavailable. Level 1 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-347 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 6 Add Cross-ties Within the Trains of the Cooling Systems -DHR
-DHRW -DHCCW Some failure combinations that eliminate both trains of the DHR related cooling systems could be mitigated if cross-ties were available between trains of the DHR, DHRW, and DHCCW systems (not between the systems). For example, these cross-ties would be helpful in conditions where the flow path fails in one train while a pump failure or maintenance event disables the opposite train. To ensure the DHR cross-ties can be implemented in a timely manner for LPI requirements, the associated valves should be operable from the main control room. Level 1 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial". 7 Use Fire Service Water as an Alternate Cooling Source for the ICCW Heat Exchangers For cases in which NSRW is unavailable due to hardware failures (e.g., flow diversion), the Fire Service Water system could be used to directly cool the ICCW heat exchangers for thermal barrier cooling support. Given that the ICCW pumps would be available for the relevant cases, a local, manual valve could be used for the alignment as time should be available for such an action. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 8 Automate Reactor Coolant Pump Trip Seal LOCAs resulting from operator failures to trip the RCPs on loss of motor bearing cooling could be reduced if high temperature sensors were installed on motor bearing cooling water lines to provide automatic trip signals. Level 1 TMI-1 Importance
 
List This SAMA's net value is positive and is classified as "cost beneficial". 9 Proceduralize Local ADV Operation TMI-1 has procedures to perform the local ADV operations that are not credited in the PRA model (the failure probability is set to 1.0). If the available procedures are credited, the RRW value of the operator action would be reduced below the SAMA review threshold. This SAMA is used demonstrate the reduction in the RRW that would occur when a reasonable failure probability is applied to the operator action. Level 1 TMI-1 Importance List Screened from analysis based on PRA insights as described in Section E.6.
: 9.
Environmental Report Appendix E  SAMA ANALYSIS Page E-348 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 10 Automate BWST Refill Failure to refill the BWST is a large contributor to some SGTR sequences, especially those in which the MS ADVs fail to operate. Automating the refill function would improve the reliability of this process and reduce the contributions from prominent SGTR sequences by providing a long term high pressure injection source. This SAMA requires a new pump with a flow rate of at least 400 gpm with a connection to a borated water source that will provide suction for 24 hours. In addition, the pump should be able to supply water from a non-borated water source for an indefinite periods of time after depletion of the borated water source. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 11 Enhance Extreme External Flooding Mitigation Equipment to Address SBO and Loss of Seal Cooling Scenarios Making the extreme flooding equipment useful for SBO conditions, especially those with TD EFW failure, would require permanently mounting the submersible pumps so that the suctions could easily be swapped from a piped water source to the flood water source. Perman ently installing the portable generator and the pumps so that they could be aligned from the MCR would improve alignm ent capabilitie s and address non-SBO loss of seal cooling cases through the ability to rapidly align alternate seal cooling. Level 1 TMI-1 Importance
 
List This SAMA's net value is positive and is classified as "cost beneficial". 12 Use the DHR System as an Alternate Suction Source for HPI Failures of the BWST suction path to the HPI pumps will lead to core damage in scenarios requiring early makeup. Through implementation of procedure changes, the DHR system could be aligned to take suction from the BWST and supply flow to the HPI system to allow injection in these cases. Level 1 TMI-1 Importance List This SAMA's net value is positive and is classified as "cost beneficial".
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-349 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 13 Change IA System Logic to Automatically Start IA-P-1A/B After a Low Voltage Trip in Conjunction with an ESAS The current IA system logic requires the operators to re-load the IA compressors on emergency power after a low voltage trip when an ESAS is registered. Automating the re-loading of these compressors would remove the requirement for the operators to perform this task in accident conditions. Level 1 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial". 14 Replace HPI Pump Cooling Alignment Valves with MOVs In the event that the normally aligned cooling source to a HPI pump fails, the current plant configuration requires local operation of the valves to swap the pump to the alternate cooling source. The time required to perform this action is considered to preclude it as a means of both preventing seal LOCAs in loss of seal cooling evolutions and for providing high pressure makeup. Replacing the valves with MOVs would allow the operators to rapidly align the alternate cooling source from the MCR in time to prevent a seal LOCA or provide high pressure injection. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 15 Automate Swap to Recirculation Mode The operator action to swap to recirculation mode is a key action for LOCA scenarios. Automating this function would
 
improve the reliability of this ac tion, especially in the rapidly evolving events where other actions are competing for the
 
attention of the operators. Level 1 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial". 16 Automate HPI Injection on Low Pressurizer Level Providing an automatic signal to initiate HPI on low pressurizer level would improve the reliability of HPI initiation. Level 1 TMI-1 Importance List This SAMA's net value is positive and is classified as "cost beneficial". 17 Auto Isolate Steam Generators on High Steam Line Flow For steam line breaks downstream of the MSIVs, failure to isolate the relevant steam generator is an important contributor to core damage. The addition of logic to isolate the steam generator on high steam line flow would reduce the core damage contribution from isolation failures. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E  SAMA ANALYSIS Page E-350 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 18 Provide the Capability to Align the Standby Battery Charger and the 1A/1B Cross-tie from the MCR TMI has a spare 125V DC battery charger for each division that can be aligned to either battery bank within a division in the event that a normally operating battery charger fails. Currently, the alignment requires local actions. There is typically adequate time to align the charger in the event of a failure, but additional changes could be made to allow rapid alignment of the spare charger from the MCR to reduce the manipulation time and improve the man-machine interface. Level 1 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 19 Install Battery Backed Hydrogen Igniters or a Passive Hydrogen Ignition System The addition of igniters would provide a means of preventing catastrophic combustible gas burns by continuously burning these gases before they reach critical levels. Providing battery backup power would increase the likelihood that this system would be available in LOOP event
: s. Use of a passive system would also function in LOOP as well as long term SBO scenarios. Level 2 TMI-1 Importance
 
List This SAMA's net value is positive and is classified as "cost beneficial". 20 Extend the High Pressure Boundary Through DHR Valve DH-V-3 for ISLOCA Isolation The highest frequency ISLOCA scenario for TMI-1 is through two valves in the DHR suction line. While the scenario's CDF is low, the release frequency is relatively high given that primary containment is bypassed by definition. No effective mitigating actions are considered to be available in these cases because 1) the break may occur upstream of DH-V-3 or additional breaks in the low pressure boundary may occur after closure of a low pressure isolation valve, 2) reduction of primary system pressure may reduce the flow out of the break, but it would not stop it, and 3) refill of the BWST does not place the plant in a stable state and results in auxiliary building flooding. Extending the pressure boundary through DH-V-3 would provide an additional isolation point in these cases. Level 2 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-351 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 21 Install Concrete Shields to Block Direct Pathways from the RPV to the Containment Wall
 
and/or Direct Containment Flooding Early in External Flooding Scenarios This SAMA is based on a failure mode identified in the Level 2 analysis that indicates corium ejection during RV failure could result in dispersal of debris such that it could directly interact with the containment wall and cause a failure of the wall. For some external flooding scenarios, it may be possible to change the procedures to direct containment flooding early such that water would be available on the containment floor before loss of power. Level 2 TMI-1 Importance
 
List This SAMA's net value is negative and is classified as "not cost beneficial". 22 Install an Independent AFW System For TMI-1, loss of MFW after a trip coupled with loss of EFW can lead to large radionuclide releases in SGTR and induced SGTR scenarios due to the unavailability of water in the SGs for fission product scrubbing. A large contributor to EFW failure is estimated to be system wide common cause failures. An independent, motor driven, auxiliary feedwater system would be an effective means of addressing these cases. Power dependence is not a large issue for the cases addressed by this SAMA and the independent EFW pump is assumed to be powered by existing emergency power such that it would not be capable of mitigating SBO scenarios. Level 2 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial". 23 Develop Alarm Response Procedures to Direct Operation of RR-V-5 on Low RBEC Flow  Failure of RR-V-6 to open results in the loss of RBEC flow to the reactor building coolers, which can be diagnosed using the system flow indicators in the main control room; however, no alarm response procedures exist to specifically direct operation
 
of the bypass valve (RR-V-5). If this procedure was developed, it may reduce the diagnosis time and improve the reliability of this operator action in an accident conditions. Level 2 TMI-1 Importance List This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E  SAMA ANALYSIS Page E-352 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 24 Install Damage Resistant, High Temperature RCP Seals with a Diesel Engine as an Alternate Drive for an EFW Pump and
 
a Portable Generator for Level Control Instrumentation For SBOs in which EFW has failed, neither primary nor secondary side cooling is available. Installing the enhanced RCP seals will prevent seal LOCAs and use of a portable generator would allow the turbine driven EFW pump to be used for extended periods in an SBO, as suggested in SAMA 2. However, in the event that the turbine driven EFW pump fails, there would be no means of providing secondary side makeup. Turbine driven EFW failures could be mitigated if an engine was available to drive one of the EFW pumps. Other industry SAMA applications have suggested similar strategies, but they typically suggest the turbine driven pumps as the best option for connection to the engine based on ease of connection. For scenarios with turbine driven EFW failure, however, the initial TD EFW pump failure may prevent its further use even with an alternate motive source. As a result, this SAMA, in addition to the requirements of SAMA 2, requires that the diesel engine be connected to one of the motor driven EFW pumps. Palisades
 
SAMA Analysis (NMC 2005) This SAMA's net value is negative and is classified as "not cost beneficial". 25 Install an Additional EDG An additional source of AC power is a potential means of supplying an entire division of safety equipment in the event that on-site AC power is lost in a LOOP. While additional EDGs are expensive, they can be cost effective at some plants, especially those with a large LOOP/SBO contribution to CDF. Palisades SAMA Analysis (NMC 2005) This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-353 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 26 Reroute Cables so that They Do Not Pass Over Ignition Sources in Fire Area CB-FA-2e (West Inverter Room) or Wrap them in Fire Proof
 
Material Some of the risk from fires in this room is from damage to cables that run over ignition sources. If the cable trays were re-routed away from the electrical equipment that they currently pass over, the consequences of equipment fires in the inverter room could be reduced.
TMI-1 IPEEE (Fire) This SAMA's net value is negative and is classified as "not cost beneficial". 27 Improve the 480V AC load center welds The IPEEE determined that the ex isting 480V AC load centers were among the weaker components in the TMI-1 AC distribution system. Adding reinforcements to the welds on the load center framework would improve the seismic durability of the structure and increase the likelihood that the system would be available after a seismic event. The other low seismic capacity components, the EDG air receivers, were enhanced subsequent to the completion of the IPEEE.
TMI-1 IPEEE (Seismic) This SAMA's net value is positive and is classified as "cost beneficial". 28 Improve the Decay Heat Service Cooler (DC-C-2A/B) Anchorages The IPEEE determined that the existing Decay Heat Service Coolers (DC-C-2A/B) lacked sufficiently durable anchorages. Replacing the anchorages with more robust anchorages would improve the seismic durability of the structure and increase the likelihood that the heat exchangers would be available after a seismic event.
TMI-1 IPEEE (Seismic) This SAMA's net value is negative and is classified as "not cost beneficial".
Environmental Report Appendix E  SAMA ANALYSIS Page E-354 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 29 Replace EDG Ground Resistors Failure of the EDG ground resistors results in failure of the EDGs, which will lead to core damage in the event that off-site power is not available. Given that the HCLPF capacity for these components was estimated at 0.25g compared with 0.09g capacities of off-site power components (such as the 1/A and 1/B distribution buses or the aux transformers), it is likely that core damage will ensue due to long term loss of power if the EDG ground resistors fail from seismic shock. Replacing the resistors with more durable versions would improve the reliability of the EDGs in seismic events.
TMI-1 IPEEE (Seismic) This SAMA's net value is negative and is classified as "not cost beneficial". 30 Improve Diesel Fire Pump Fuel Oil Tank and Battery Rack Supports The Fire Service Water system provides cooling to the SBO EDG, backup cooling the DHCCW heat exchangers, and backup cooling to the "1A" and "1B" Instrument Air compressors. While seismic failures to the systems FSW supports would likely limit the benefit of improving the fuel oil tank and battery racks, some benefit may be available through improvements to the diesel fire pump's reliability.
TMI-1 IPEEE (Fire/Seismic)This SAMA's net value is negative and is classified as "not cost beneficial". 31 Modify Specific Containment Penetration MOVs to "Fail Closed" Most containment penetrations have AOV or SOV isolation valves that will fail closed on loss of air or power; however, there are cases in which MOVs are used instead. Those lines that do not include a pair of AOVs or SOVs that fail closed are typically below 1" in diameter or include at least one AOV or SOV that will fail closed on loss of air or power. However, the NSCCW and RBEC systems include penetrations that only include MOVs. While these are closed cooling systems that would not normally provide a credible release path, heat exchanger breaks in seismic events could provide containment bypass routes in the event that a failure also occurs in the reactor building. Changing one of the valves in each of these paths to fail closed is a means of increasing the isolation probability over what is available from manual action.
TMI-1 IPEEE (Seismic) Screened from analysis based on PRA insights as described in Section E.6.
3 1.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-355 License Renewal Application TABLE E.5-4 PHASE II SAMA SAMA NUMBER SAMA TITLE SAMA DESCRIPTION SOURCE BASELINE PHASE II DISPOSITION 32 Pre-stage Severe Flooding Equipment Pre-staging the equipment used to prevent core damage in severe flooding conditions would reduce sources of error in the alignment actions and reduce the time required to perform the task. Potential changes include: - Storing the portable EDG on the turbine deck
- Adding a normally empty fuel oil tank for the portable EDG to the turbine deck - Permanently running power cable from the portable EDG to the pump areas A potential permutation of this SAMA would be to procure an additional portable EDG to reduce the failure contribution from the power source.
TMI-1 IPEEE (External Flooding) This SAMA's net value is positive and is classified as "cost beneficial". 33 Increase the Flood Protection Height The current configuration protects to the design basis limit of 310 feet msl and levels any higher result in topping of the existing flood doors and flooding of sensitive areas. Raising the height of the flood doors (or completely sealing the doors) would prevent water incursion and allow for continued operation of the normal safety equipment.
TMI-1 IPEEE (External Flooding) This SAMA's net value is positive and is classified as "cost beneficial".
Environmental Report Appendix E  SAMA ANALYSIS Page E-356 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 8 Automate Reactor Coolant Pump Trip $145,000 $3,395,359 $3,250,35923.4 This SAMA would complement the set of existing RCP protection signals to protect against potential cooling failures that appear to be critical to the RCPs. Given the relatively low implementation cost and the relatively large risk reduction associated with the change, this SAMA is a candidate for implementation. SAMA 32 Pre-stage Severe Flooding Equipment $1,700,000 $35,893,061$34,193,06121.1 This SAMA yields a large averted cost-risk for TMI-1. There is a large degree of uncertainty associated with flood risk that could impact the results of the cost benefit analysis, but the location of the plant suggests that enhancements to the extreme flood mitigation strategy should be in place for the site. This SAMA should be considered for implementation. SAMA 19 Install Battery Backed Hydrogen Igniters or a Passive Hydrogen Ignition System $760,000 $8,601,659 $7,841,65911.3 The passive hydrogen ignition system is designed to prevent containment failures due to post-core-damage combustible gas burns in accident conditions and is intended to be operable even in long term SBO evolutions. The current PRA model considers combustible gas burns to be a credible containment failure mode, but the conservative assumptions related to the containment failure probabilities are considered to greatly overestimate the benefit of this SAMA. This SAMA is not recommended for implementation. SAMA 12 Use the DHR System as an Alternate Suction Source for HPI $50,000 $545,705 $495,705 10.9 This is an inexpensive change that would allow the operators to use HPI in the event that the normal BWST suction path fails. While the probability that the alternate suction alignment would be required during the life of the plant is low, this SAMA would proceduralize a means of addressing failures that could otherwise contribute to core damage. This SAMA should be considered for implementation.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-357 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 11 Enhance Extreme External Flooding Mitigation Equipment to Address SBO and Loss of Seal Cooling Scenarios $4,250,000 $44,243,903$39,993,90310.4 SAMA 11 is a complex plant modification that was designed to reduce internal events SBO risk by taking advantage of equipment that could also be used to mitigate the extreme flood scenarios. The intent of the SAMA was to determine if changes could be made to the extreme flooding equipment such that it would be beneficial in non-external flooding SBO cases. However, the differences in the external flooding SBO and a standard SBO require significantly different capabilities. The main issue is that the external flooding strategy uses the flood cues to predict the need for the mitigation equipment well before the loss of AC power. The implication is that seal cooling can be maintained such that there will not be a seal LOCA.
SAMA 11 (cont.)      If a seal LOCA did occur, the primary side makeup requirements increase and the injection inventory may be depleted over the long potential mission times for external flooding events. Consequently, seal LOCA prevention is considered to be a requirement for long term success. For standard SBO cases, seal LOCAs are assumed to be preventable only if seal cooling can be restored within 13 minutes of the initial loss of cooling (standard SBOs are generally not anticipated and the mitigation equipment could not be pre-initiated). Seal LOCA prevention would require an auto start/load of the 480V AC generator on an undervoltage signal to the HPI pump buses or high RCP cooling water temperature signal. Even without external flooding contributions, this SAMA would be cost beneficial based on the 95th percentile PRA results. However, SAMA 2 may be a more desirable means of addressing seal LOCAs given that its passive design would likely be more reliable than an active cooling system and because it yields a larger internal events risk reduction, which has benefits outside of the SAMA analysis.
Environmental Report Appendix E  SAMA ANALYSIS Page E-358 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 33 Increase the Flood Protection Height $2,700,000 $25,141,284$22,441,2849.3 This SAMA is a potential means of mitigating severe flood risk; however, this strategy is predicated on identifying and eliminating all flow paths into areas containing safety equipment. In addition, there is the implicit assumption that the flood gates and buildings will withstand the hydrodynamic forces of the flood waters. Because of the uncertainty associated with this SAMA, SAMA 32 is considered to be the better approach to addressing flood risk and SAMA 33 a less desirable alternative. If SAMA 32 is implemented, SAMA 33 would not be cost beneficial. SAMA 27 Improve the 480V AC load center welds $575,000 $3,593,752 $3,018,7526.3 This modification was identified in the IPEEE as a change that could reduce seismic risk by about 50 percent. While this enhancement addressed a significant seismic concern, the modifications were not implemented because the load center failures only accounted for about 10 percent of the total TMI-1 CDF (internal + external events). If the LLNL seismic hazard curves are used in place of the EPRI seismic hazard curves that were used in the IPEEE base case, the seismic CDF increases from 3.21E-05/yr to 8.43E-05/yr. Given this condition, strengthening the 480 V AC load center welds would yield a CDF reduction of 4.22E-05/yr. While this appears to be a likely candidate for implementation, the seismic hazard curves represent a source of uncertainty in the seismic risk evaluation. Because TMI-1 is located in a seismically stable region, this SAMA may warrant further review, but it is not suggested for implementation at this time.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-359 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 16 Automate HPI Injection on Low Pressurizer Level $1,100,000 $4,379,735 $3,279,7354.0 This SAMA suggests further automating an action on which the operators are well trained. While operator training is thorough and manual initiation failures are very unlikely even in cases where the current initiation logic would not actuate, failure to manually initiate HPI implies that a severe diagnosis error has occurred and that subsequent actions are also in jeopardy of failing. Even though the automatic HPI initiation could be inhibited/cancelled, such an action would require an active assessment of the RCS level and it would provide an opportunity for level recovery. However, the benefit of this SAMA is based on the PRA human error probability assessments, which are typically associated with a relatively high degree of uncertainty. In this case, a single joint human error probability is responsible for most of the PRA model-predicted risk and it is not appropriate to justify a large expenditure of resources to address a risk area with such a wide uncertainty. This SAMA should not be considered as a high priority item. SAMA 21 Install Concrete Shields to Block Direct Pathways from the RPV to the Containment Wall and/or Direct Containment Flooding Early in External Flooding Scenarios $1,200,000 $3,248,127 $2,048,1272.7 This SAMA yields a relatively low benefit for internal events even for the 95th percentile results (about $560k), but when the benefits associated with the external flood contributions are added, the SAMA shows a much higher benefit. Implementation of SAMA 32 would reduce the benefit associated with this SAMA to the point where it would no longer be cost beneficial. If SAMA 32 is implemented, SAMA 21 should not be considered for implementation. Even without implementation of SAMA 32, discussions with Severe Accident Management personnel indicate that the path from the reactor vessel to the containment shell is obstructed and that the shell liner failure probability used in the PRA may be pessimistic. SAMA 23 Develop Alarm Response Procedures to Direct Operation of RR-V-5 on Low RBEC Flow  $50,000 $84,230 $34,230 1.7 SAMA 23 is a low cost procedure change that would help the operators diagnose containment cooling problems. This SAMA should be considered for implementation.
Environmental Report Appendix E  SAMA ANALYSIS Page E-360 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 2 Install Damage Resistant, High Temperature RCP Seals with a Portable 480V Generator for Extended EFW Operation $7,300,000 $11,816,753$4,516,7531.6 SAMA 2 is a high cost change that impacts seal LOCAs and SBO scenarios. When the 95th percentile PRA results are considered, this SAMA is shown to be potentially cost beneficial. While the DPD ratio is smaller than what has been estimated for SAMA 11, SAMA 2 may be a more desirable means of addressing seal LOCAs given that its passive design would likely be more reliable than an active cooling system and because it yields a larger internal events risk reduction, which has benefits outside of the SAMA analysis. SAMA 24 Install Damage Resistant, High Temperature RCP Seals with a Diesel Engine as an Alternate Drive for an EFW Pump and a Portable Generator for Level Control Instrumentation $8,400,000 $12,144,553$3,744,5531.4 SAMA 24 is an enhancement of SAMA 2 that is designed to address turbine driven EFW failures. Given that the difference in benefit between SAMA 2 and SAMA 24 when considering the 95th percentile PRA results is only $327,800, it would not be beneficial to add the diesel driven motor option for the EFW pump when the cost of that portion of the SAMA is estimated to be $1.1 million. This SAMA is not recommended for implementation.
SAMA 7 Use Fire Service Water as an Alternate Cooling Source for the ICCW Heat Exchangers $1,000,000 $1,235,449 $235,449 1.2 SAMA 7 provides an alternate means of cooling the ICCW heat exchangers when normal cooling flow to the heat exchangers fails.
While this enhancement provides a non-negligible reduction in risk, the margin by which it is cost beneficial is low and it is not a likely candidate for implementation. SAMA 15 Automate Swap to Recirculation Mode $450,000 $547,520 $97,520 1.2 SAMA 15 is a SAMA that has been identified for many plants in the industry. For TMI-1, it is only considered to be cost effective using the 95th percentile PRA results and a generic implementation cost of $450,000, which may be low. This is not a high priority candidate for implementation based on the small margin by which it is cost effective and because aplant specific implementation cost estimate may provide a basis for excluding it from consideration.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-361 License Renewal Application TABLE E.8-1 SUMMAY OF COST BENEFICIAL SAMAS SAMA ID SAMA Title SAMA Implementation Cost Averted Cost-Risk Net ValueDPD Ratio*Comments SAMA 26 Reroute Cables so that They Do Not Pass Over Ignition Sources in Fire Area CB-FA-2e (West Inverter Room) or Wrap them in Fire Proof Material $900,000 $1,016,573 $116,573 1.1 The margin by which this SAMA is cost beneficial is small and the methods available to estimate the averted cost-risk were limited, as
 
descr i bed i n Section E.5.1.6
.1. T h is SAMA m a y b e con s i d ered cost beneficial, but a more detailed, up to date assessment of the fire risk would be required to better define the potential benefit of protecting the cables in Fire Area CB-FA-2E. Table Notes:
* The DPD (dollar per dollar) Ratio is the Averted Cost-Risk divided by the SAMA cost. ** The absolute change in CDF (baseline CDF minus estimated CDF with the particular SAMA in place) is presented followed by the percent change (in parentheses).
Environmental Report Appendix E  SAMA ANALYSIS Page E-362 Three Mile Island Nuclear Station Unit 1 License Renewal Application E.11 REFERENCES ABS 2000 ABS Consulting, Three Mile Island Unit 1 Probabilistic Risk Assessment (Level 1 Update, August 2000. ABS 2003 ABS Consulting, TMI 2003 Update , Job No. 1194243, August 2003. BGE 1998 BGE (Baltimore Gas and Electric). 1998. Calvert Cliffs Application for License Renewal , Attachment 2 of Appendix F - Severe Accident Mitigation Alternatives Analysis. April.
Chanin and Young  1997 Chanin, David I. and Mary L. Young. 1997. Code Manual for MACCS2:  Volume 1, User's Guide, SAND97-0594, Sandia National Laboratories, Albuquerque, New Mexico. March. CPL 2002 CPL (Carolina Power and Light). 2002. Applicant's Environmental Report; Operating License Renewal Stage; H. B. Robinson Steam Electric Plant Unit No. 2. Appendix F Severe Accident Mitigation Alternatives, Letter, J. W. Moyer (CP&L) to U.S. Nuclear Regulatory Commission.  "Application for Renewal of Operating License."  June
: 14. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/ro binson.html.
CPL 2004 CPL (Carolina Power and Light). 2004. Applicant's Environmental Report; Operating License Renewal Stage; Brunswick Steam Electric Plant. Appendix F Severe Accident Mitigation Alternatives. October. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/br unswick.html.
DE&S 1992 DE&S (Duke Engineering & Services). 1992. Discussion of LOCA break size criteria based on Level 1 PRA results. Telecon between David Noxon (DE&S) and Chuck Adams (GPUN). DE&S File No: MTS-4042.
May 26.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-363 License Renewal Application Duke 1990 Duke (Duke Power Company). 1990. Oconee Nuclear Station Unit 3 PRA, R e vision 1, Secti o ns:  2.3, 6.0, 6.1, 6.2, Appendix A.3. February
: 26. Duke 1998 Duke (Duke Energy Corporation). 1998. Applicant's Environmental Report; Operating License Renewal Stage; Oconee Nuclear Station Units 1, 2, and 3. Attachment K Severe Accident Mitigation Alternatives. April. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/ocon ee/exhibitd.pdf Duke 2001 Duke (Duke Energy Corporation). 2001. Applicant's Environmental Report; Operating License Renewal Stage; Catawba Nuclear Station. Attachment H Severe Accident Mitigation Alternatives. May. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/mcgu ire-catawba/duke-env-c-att2.pdf EPA 1972 EPA (U.S. Environmental Protection Agency). 1972. Mixing Heights, Wind Speeds, and Potential for Urban Air Pollution Throughout the Contiguous United States. AP-101. Holzworth. George. C. January. EPRI 1989 EPRI (Electric Power Research Institute). 1989. Probabilistic Seismic Hazard Evaluation at Nuclear Plant Sites in the Central and Eastern United States: Resolution of the Charleston Issue. April. EPRI 1995 EPRI (Electric Power Research institute). 1995.
PSA Applications Guide. EPRI TR-105396, Final Report. D.E. True. August. ETI 2003 ETI (Earth Tech, Inc). 2003. Evacuation Time Estimates for the Three Mile Island Station Emergency Planning Zone. July.
Environmental Report Appendix E  SAMA ANALYSIS Page E-364 Three Mile Island Nuclear Station Unit 1 License Renewal Application Exelon 2001 EXELON (Exelon Corporation). 2001. Peach Bottom Application for License Renewal, PBAPS (Peach Bottom Atomic Power Station). Appendix E - Environmental Report and Appendix G - Severe Accident Mitigation Alternatives. Exelon 2003 EXELON (Exelon Corporation). 2003. Applicant's Environmental Report; Operating License Renewal Stage; Quad Cities Nuclear Power Station Units 1 and 2. Section 4.20 Severe Accident Mitigation Alternatives (SAMA) and Appendix F SAMA Analysis, Letter, Benjamin, Exelon, to U. S. Nuclear Regulatory Commission. Application for Renewed Operating Licenses. January 3. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/dresden-quad.html Exelon 2005a Exelon (Exelon Corporation). 2005a. Three Mile Island PRA Notebook, "Event Tree Analysis", TMI-PRA-001, Revision 0. June. Exelon 2005b Exelon (Exelon Corporation). 2005b. TMI PRA Notebook, TMI-PRA-014 Quantification Notebook, TMI PRA Model 2004 Revision 1, June. Exelon 2006a Exelon (Exelon Corporation). 2006a. TMI-1 System Notebook, "Reactor Building Spray System Notebook", P0467050015-2538, TM-PRA-
 
010.17 RBSS, March. Exelon 2006b Exelon (Exelon Corporation). 2006b. TMI-1 System Notebook, "Reactor Building Emergency Cooling System Analysis System Notebook",
P0467050015-2523, TM-PRA-010.16 RBECS, March. Exelon 2006c Exelon (Exelon Corporation). 2006c. TMI-1 System Notebook, "Reactor Building Isolation System Notebook", P0467050015-2545, TM-PRA-010.15 RBIS, March. Exelon 2007a Exelon (Exelon Corporation). 2007a. TMI PRA Notebook, TMI-PRA-014 Quantification Notebook, TMI PRA Model 2004 Revision 2, May.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-365 License Renewal Application Exelon 2007b Exelon (Exelon Corporation). 2007b. TMI PRA Notebook, TMI-PRA-015.2 Level 2 Containment Event Tree Analysis Notebook , TMI PRA Model 2004 Revision 2, May. Exelon 2007c Exelon (Exelon Corporation). 2007c. Order of Magnitude Cost Estimates for TMI SAMAs. E-mail from David Honan (Exelon) to J.R. Gabor (ERIN). May 31 st. FAI 1987 FAI (Fauske and Associates, Inc.). 1987. Modular Accident Analysis Program (MAAP) User's Manual. IDCOR Technical Report 16.2-3.
1987. FPL 2000 FPL (Florida Power & Light). 2000. Application for Renewed Operating Licenses; Turkey Point Units 3 & 4. Appendix F Severe Accident Mitigation Alternatives Analysis. September. Available on U. S.
 
Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/tu rkey-point/lra.pdf GPU 1990 GPU (GPU Nuclear Corporation). 1990. External Flooding at TMI-1. Memorandum from P. S. Brady (GPUN) to J. S. Wetmore (GPUN). TMINS 5523-90-121. May 08. GPU 1992 GPU (GPU Nuclear Corporation) and PLG, Inc. 1992.
TMI Unit 1 Probabilistic Risk Assessment (Level 1) Update. December. GPU 1993a GPU (GPU Nuclear Corporation). 1993a. Duke Engineering Services, and B&W Nuclear Services, "TMI-1 Probabilistic Risk Assessment (Level 2)", January 1993. GPU 1993a GPU (GPU Nuclear Corporation). 1993a. TMI-1 IPE Submittal Report , March. GPU 1993b GPU (GPU Nuclear Corporation). 1993b. Duke Engineering Services, and B & W Nuclear Services, "TMI-1 Probabilistic Risk Assessment (Level 2)", February 1993.
Environmental Report Appendix E  SAMA ANALYSIS Page E-366 Three Mile Island Nuclear Station Unit 1 License Renewal Application GPU 1993b GPU (GPU Nuclear Corporation), Duke Engineering Services, and B&W Nuclear Services. 1993b. TMI Unit 1 Probabilistic Risk Assessment (Level 2). January. GPU 1994 GPU (GPU Nuclear Corporation). 1994. TMI Unit 1 Individual Plant Examination for External Events Submittal Report, December. IEM 2002 IEM (Innovative Emergency Management). 2002. Evacuation Time Estimates for the Harris Nuclear Plant. Final Report. December 29. NC 2005 NC (State of North Carolina). 2005. "Population Overview: 2000-2030." State Demographics Unit. May 25. Available at http://demog.state.nc.us/demog/pop0030.xls (html and excel format).
Accessed April 4, 2006. NMC 2004 NMC (Nuclear Management Company, LLC). 2004.
Applicant's Environmental Report; Operating License Renewal Stage; Point Beach Nuclear Plant, Units 1 and 2. Appendix F SAMA Analysis. Application for Renewed Operating Licenses. February. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/point-beach/er.pdf NMC 2005 NMC (Nuclear Management Company, LLC). 2005.
Applicant's Environmental Report; Operating License Renewal Stage; Palisades
 
Nuclear Plant. Attachment E SAMA Analysis. Application for
 
Renewed Operating Licenses. March. Available on U. S. Nuclear Regulatory Commission website at http://www.nrc.gov/reactors/operating/licensing/renewal/applications/p alisades/palisades_er.pdf NRC 1976 NRC (U.S. Nuclear Regulatory Commission). 1976.
Flood Protection for Nuclear Power Plants.
Regulatory Guide 1.102. Office of Standards Development. Washington, D.C., September.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-367 License Renewal Application NRC 1977 NRC (U.S. Nuclear Regulatory Commission). 1977. Design Basis Floods for Nuclear Power Plants. Regulatory Guide 1.59. Office of Standards Development. Washington, D.C., August. NRC 1982 NRC (U.S. Nuclear Regulatory Commission). 1982. Analysis of Hypothetical Severe Core Damage Accidents for the Zion Pressurized Water Reactor. NUREG/CR-1989. Washington, DC. October. NRC 1983 NRC (U.S. Nuclear Regulatory Commission). 1983. Handbook of Human Reliability with Emphasis on Nuclear Power Plant Applications. Final Report. NUREG/CR-1278. Washington, DC. August. NRC 1987 NRC (U.S. Nuclear Regulatory Commission). 1987.
Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants. LWR Edition. NUREG-0800. Washington, D.C., June. NRC 1990 NRC (U.S. Nuclear Regulatory Commission). 1990. Evaluation of Severe Accident Risks: Quantification of Major Input Parameters, NUREG/CR-4551, SAND86-1309, Vol. 2, Rev. 1, Part 7, Sprung, J.L., Rollstin, J.A., Helton, J.C., Jow, H-N. Washington, DC. December. NRC 1991 NRC (U.S. Nuclear Regulatory Commission). 1991. Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities - 10 CFR 50.54(f). Generic Letter 88-20, Supplement
: 4. April. NRC 1994 NRC (U.S. Nuclear Regulatory Commission). 1994. Revised Livermore Seismic Hazard Estimates for 69 Nuclear Power Plant Sites East of the Rocky Mountains. NUREG-1488. NRC 1997 NRC (U.S. Nuclear Regulatory Commission). 1997.
Regulatory Analysis Technical Evaluation Handbook. NUREG/BR-0184. NRC 1998a NRC (U.S. Nuclear Regulatory Commission). 1998a. Code Manual for MACCS2: User's-Guide. NUREG/CR-6613, Volume 1, SAND 97-
 
0594. Chanin, D. and Young, M. May.
Environmental Report Appendix E  SAMA ANALYSIS Page E-368 Three Mile Island Nuclear Station Unit 1 License Renewal Application NRC 1998b NRC (U.S. Nuclear Regulatory Commission). 1998b. Common-Cause Failure Parameter Estimations, Marshall, F.M., Rasmuson, D. M. and Mosleh, A. NUREG/CR-5497, October 1998 (data updated through December 31, 2001 in the electronic file CCFParamEst.chm dated 6/24/2003 provided by T. Wierman of INEEL, March 2004). NRC 2003 NRC (U.S. Nuclear Regulatory Commission). 2003. Sector Population, Land Fraction, and Economic Estimation Program. SECPOP2000:
NUREG/CR-6525, Washington, D.C., Rev. 1, August. PPL  2006 PPL (PPL Susquehanna, LLC). 2006. Susquehanna Steam Electric Station Application for License Renewal. Environmental Report.
Appendix E. August. PRC 1981 PRC (PRC Voorhees Company). 1981. Oconee Nuclear Station Evacuation Analysis, Evacuation Time Estimates, December. SCE&GC 2002 SCE&GC (South Carolina Electric and Gas Company). 2002. Virgil C. Summer Nuclear Station Application for License Renewal. Environmental Report. Appendix F. August. SNOC 2000 SNOC (Southern Nuclear Operating Company). 2000. Edwin I. Hatch Nuclear Plant Application for License Renewal, Environmental Report.
Appendix D, Attachment F. February. TVA 2003 TVA (Tennessee Valley Authority). 2003. Applicant's Environmental Report; Operating License Renewal Stage; Browns Ferry Nuclear Power Plant, Units 1, 2, and 3. Attachment E-4,  Severe Accident Mitigation Alternatives at the Browns Ferry Nuclear Plant, Volume I of III. Application for Renewed Operating Licenses. December. USDA 1998 USDA (U.S. Department of Agriculture). 1998. 1997 Census of Agriculture. National Agricultural Statistics Service. http://www.nass.usda.gov/census/census97/volume1/ vol1pubs.htm
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-369 License Renewal Application ADDENDUM 1 TO ATTACHMENT E SELECTED PREVIOUS INDUSTRY SAMAS
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-370 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement Improvements Related to RCP Seal LOCAs (Loss of CC or SW) 1 Cap downstream piping of normally closed component cooling water drain and vent valves.
SAMA would reduce the frequency of a loss of component cooling event, a large portion of which was derived from catastrophic failure of one of the many single isolation valves.
2 Enhance loss of component cooling procedure to facilitate stopping reactor coolant pumps.
SAMA would reduce the potential for reactor coolant pump (RCP) seal damage due to pump bearing failure.
3 Enhance loss of component cooling procedure to present desirability of cooling down reactor coolant system (RCS) prior to seal LOCA.
SAMA would reduce the potential for RCP seal failure.
4 Provide additional training on the loss of component cooling.
SAMA would potentially improve the success rate of operator actions after a loss of component cooling (to restore RCP seal damage).
5 Provide hardware connections to allow another essential raw cooling water system to cool charging pump seals.
SAMA would reduce effect of loss of component cooling by providing a means to maintain the centrifugal charging pump seal injection after a loss of component cooling.
6 Procedure changes to allow cross connection of motor cooling for residual heat removal service water (RHRSW) pumps.
SAMA would allow continued operation of both RHRSW pumps on a failure of one train of PSW.
7 Proceduralize shedding component cooling water loads to extend component cooling heatup on loss of essential raw cooling water.
SAMA would increase time before the loss of component cooling (and reactor coolant pump seal failure) in the loss of essential raw cooling water sequences.
8 Increase charging pump lube oil capacity.
SAMA would lengthen the time before centrifugal charging pump failure due to lube oil overheating in loss of CC sequences.
9 Eliminate the RCP thermal barrier dependence on component cooling such that loss of component cooling does not result directly in core damage.
SAMA would prevent the loss of recirculation pump seal integrity after a loss of component cooling. Watts Bar Nuclear Plant IPE said that they could do this with essential raw cooling water connection to RCP seals.
10 Add redundant DC control power for PSW pumps C & D.
SAMA would increase reliability of PSW and decrease CDF due to a loss of SW. 11 Create an independent RCP seal injection system, with a dedicated diesel.
SAMA would add redundancy to RCP seal cooling alternatives, reducing CDF from loss of component cooling or SW or from a SBO event.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-371 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 12 Use existing hydro-test pump for RCP seal injection.
SAMA would provide an independent seal injection source, without the cost of a new system.
13 Replace ECCS pump motor with air-cooled motors.
SAMA would eliminate ECCS dependency on component cooling system (but not on room cooling).
14 Install improved RCS pumps seals.
SAMA would reduce probability of RCP seal LOCA by installing RCP seal O-ring constructed of improved materials 15 Install additional component cooling water pump.
SAMA would reduce probability of loss of component cooling leading to RCP seal LOCA.
16 Prevent centrifugal charging pump flow diversion from the relief valves.
SAMA modification would reduce the frequency of the loss of RCP seal cooling if relief valve opening causes a flow diversion large enough to prevent RCP seal injection.
17 Change procedures to isolate RCP seal letdown flow on loss of component cooling, and guidance on loss of injection during seal LOCA.
SAMA would reduce CDF from loss of seal cooling.
18 Implement procedures to stagger high-pressure safety injection (HPSI) pump use after a loss of SW.
SAMA would allow HPSI to be extended after a loss of SW.
19 Use FPS pumps as a backup seal injection and high-pressure
 
makeup. SAMA would reduce the frequency of the RCP seal LOCA and the SBO CDF. 20 Enhance procedural guidance for use of cross-tied component cooling or SW pumps.
SAMA would reduce the frequency of the loss of component cooling water and SW. 21 Procedure enhancements and operator training in support system failure sequences, with emphasis on anticipating problems and coping.
SAMA would potentially improve the success rate of operator actions subsequent to support system failures.
22 Improved ability to cool the residual heat removal (RHR) heat exchangers.
SAMA would reduce the probability of a loss of decay heat removal by implementing procedure and hardware modifications to allow manual alignment of the FPS or by installing a component cooling water cross-tie.
23 Additional SW Pump SAMA would conceivably reduce common cause dependencies from SW system and thus reduce plant risk through system reliability improvement.
24 Create an independent RCP seal injection system, without dedicated diesel This SAMA would add redundancy to RCP seal cooling alternatives, reducing the CDF from loss of CC or SW, but not SBO.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-372 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement Improvements Related to Heating, Ventilation, and Air Conditioning 25 Provide reliable power to control building fans.
SAMA would increase availability of CR ventilation on a loss of power.
26 Provide a redundant train of ventilation.
SAMA would increase the availability of components dependent on room cooling. 27 Procedures for actions on loss of HVAC.
SAMA would provide for improved credit to be taken for loss of HVAC sequences (improved affected electrical equipment reliability upon a loss of control building HVAC).
28 Add a diesel building switchgear room high temperature alarm.
SAMA would improve diagnosis of a loss of switchgear room HVAC.
Option 1:  Install high temp alarm. Option 2:  Redundant louver and thermostat 29 Create ability to switch fan power supply to DC in an SBO event. SAMA would allow continued operation in an SBO event. This SAMA was created for reactor core isolation cooling (RCIC) system room at Fitzpatrick Nuclear Power Plant.
30 Enhance procedure to instruct operators to trip unneeded RHR/CS pumps on loss of room ventilation.
SAMA increases availability of required RHR/CS pumps. Reduction in room heat load allows continued operation of required RHR/CS pumps, when room cooling is lost.
31 Stage backup fans in switchgear (SWGR) rooms This SAMA would provide alternate ventilation in the event of a loss of SWGR Room ventilation Improvements Related to Ex-Vessel Accident Mitigation/Containment Phenomena 32 Delay containment spray actuation after large LOCA.
SAMA would lengthen time of refueling water storage tank (RWST) availability.
33 Install containment spray pump header automatic throttle valves. SAMA would extend the time over which water remains in the RWST, when full CS flow is not needed 34 Install an independent method of suppression pool cooling.
SAMA would decrease the probability of loss of containment heat removal. For PWRs, a potential similar enhancement would be to install an independent cooling system for sump water.
35 Develop an enhanced drywell spray system.
SAMA would provide a redundant source of water to the containment to control containment pressure, when used in conjunction with containment heat removal.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-373 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 36 Provide dedicated existing drywell spray system.
SAMA would provide a source of water to the containment to control containment pressure, when used in conjunction with containment heat removal. This would use an existing spray loop instead of developing a new spray system.
37 Install an unfiltered hardened containment vent.
SAMA would provide an alternate decay heat removal method for non-ATWS events, with the released fission products not being scrubbed.
38 Install a filtered containment vent to remove decay heat.
SAMA would provide an alternate decay heat removal method for non-ATWS events, with the released fission products being scrubbed. Option 1:  Gravel Bed Filter Option 2:  Multiple Venturi Scrubber 39 Install a containment vent large enough to remove ATWS decay heat.
Assuming that injection is available, this SAMA would provide alternate decay heat removal in an ATWS event.
40 Create/enhance hydrogen recombiners with independent power supply.
SAMA would reduce hydrogen detonation at lower cost,  Use either 1) a new independent power supply 2) a nonsafety-grade portable generator 3) existing station batteries
: 4) existing AC/DC independent power supplies.
41 Install hydrogen recombiners.
SAMA would provide a means to reduce the chance of hydrogen detonation.
42 Create a passive design hydrogen ignition system.
SAMA would reduce hydrogen denotation system without requiring electric power. 43 Create a large concrete crucible with heat removal potential under the basemat to contain molten core debris.
SAMA would ensure that molten core debris escaping from the vessel would be contained within the crucible. The water cooling mechanism would cool the molten core, preventing a melt-through of the basemat.
44 Create a water-cooled rubble bed on the pedestal.
SAMA would contain molten core debris dropping on to the pedestal and would allow the debris to be cooled.
45 Provide modification for flooding the drywell head.
SAMA would help mitigate accidents that result in the leakage through the drywell head seal.
46 Enhance FPS and/or standby gas treatment system hardware and procedures.
SAMA would improve fission product scrubbing in severe accidents.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-374 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 47 Create a reactor CFS.
SAMA would enhance debris coolability, reduce core concrete interaction, and provide fission product scrubbing.
48 Create other options for reactor cavity flooding.
SAMA would enhance debris coolability, reduce core concrete interaction, and provide fission product scrubbing.
49 Enhance air return fans (ice condenser plants).
SAMA would provide an independent power supply for the air return fans, reducing containment failure in SBO sequences.
50 Create a core melt source reduction system.
SAMA would provide cooling and containment of  molten core debris.
Refractory material would be placed underneath the reactor vessel such that a molten core falling on the material would melt and combine with the material. Subsequent spreading and heat removal form the vitrified compound would be facilitated, and concrete attack would not occur 51 Provide a containment inerting capability.
SAMA would prevent combustion of hydrogen and carbon monoxide gases.
52 Use the FPS as a backup source for the containment spray system. SAMA would provide redundant containment spray function without the cost of installing a new system.
53 Install a secondary containment filtered vent.
SAMA would filter fission products released from primary containment.
54 Install a passive containment spray system.
SAMA would provide redundant containment spray method without high
 
cost. 55 Strengthen primary/secondary containment.
SAMA would reduce the probability of containment overpressurization to failure. 56 Increase the depth of the concrete basemat or use an alternative concrete material to ensure melt-through does not occur. SAMA would prevent basemat melt-through.
57 Provide a reactor vessel exterior cooling system.
SAMA would provide the potential to cool a molten core before it causes vessel failure, if the lower head could be submerged in water.
58 Construct a building to be connected to primary/secondary containment that is maintained at a vacuum.
SAMA would provide a method to depressurize containment and reduce fission product release.
59 Refill CST SAMA would reduce the risk of core damage during events such as extended SBOs or LOCAs which render the suppression pool unavailable as an injection source due to heat up.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-375 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 60 Maintain ECCS suction on CST SAMA would maintain suction on the CST as long as possible to avoid pump failure as a result of high suppression pool temperature 61 Modify containment flooding procedure to restrict flooding to below TAF SAMA would avoid forcing containment venting 62 Enhance containment venting procedures with respect to timing, path selection and technique.
SAMA would improve likelihood of successful venting strategies.
63 Severe Accident EPGs/AMGs SAMA would lead to improved arrest of core melt progress and prevention of containment failure 64 Simulator Training for Severe Accident SAMA would lead to improved arrest of core melt progress and prevention of containment failure 65 Dedicated Suppression Pool Cooling SAMA would decrease the probability of loss of containment heat removal.
While PWRs do not have suppression pools, a similar modification may be applied to the sump. Installation of a dedicated sump cooling system would provide an alternate method of cooling injection water.
66 Larger Volume Containment SAMA increases time before containment failure and increases time for recovery 67 Increased Containment Pressure Capability (sufficient pressure to withstand severe accidents)
SAMA minimizes likelihood of large releases 68 Improved Vacuum Breakers (redundant valves in each line)
SAMA reduces the probability of a stuck open vacuum breaker.
69 Increased Temperature Margin for Seals This SAMA would reduce containment failure due to drywell head seal failure caused by elevated temperature and pressure.
70 Improved Leak Detection This SAMA would help prevent LOCA events by identifying pipes which have begun to leak. These pipes can be replaced before they break.
71 Suppression Pool Scrubbing Directing releases through the suppression pool will reduce the radionuclides allowed to escape to the environment.
72 Improved Bottom Penetration Design SAMA reduces failure likelihood of RPV bottom head penetrations 73 Larger Volume Suppression Pool (double effective liquid volume) SAMA would increase the size of the suppression pool so that heatup rate is reduced, allowing more time for recovery of a heat removal system
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-376 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 74 Unfiltered Vent SAMA would provide an alternate decay heat removal method with the released fission products not being scrubbed.
75 Filtered Vent SAMA would provide an alternate decay heat removal method with the released fission products being scrubbed.
76 Post Accident Inerting System SAMA would reduce likelihood of gas combustion inside containment 77 Hydrogen Control by Venting Prevents hydrogen detonation by venting the containment before combustible levels are reached.
78 Pre-inerting SAMA would reduce likelihood of gas combustion inside containment 79 Ignition Systems Burning combustible gases before they reach a level which could cause a harmful detonation is a method of preventing containment failure.
80 Fire Suppression System Inerting Use of the FPS as a back up containment inerting system would reduce the probability of combustible gas accumulation. This would reduce the containment failure probability for small containments (e.g. BWR MKI).
81 Drywell Head Flooding SAMA would provide intentional flooding of the upper drywell head such that if high drywell temperatures occurred, the drywell head seal would not fail. 82 Containment Spray Augmentation This SAMA would provide additional means of providing flow to the containment spray system.
83 Integral Basemat This SAMA would improve containment and system survivability for seismic events. 84 Reactor Building Sprays This SAMA provides the capability to use firewater sprays in the reactor building to mitigate release of fission products into the Rx Bldg following an accident. 85 Flooded Rubble Bed SAMA would contain molten core debris dropping on to the pedestal and would allow the debris to be cooled.
86 Reactor Cavity Flooder SAMA would enhance debris coolability, reduce core concrete interaction, and provide fission product scrubbing.
87 Basaltic Cements SAMA minimizes carbon dioxide production during core concrete interaction.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-377 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 88 Provide a core debris control system (Intended for ice condenser plants): This SAMA would prevent the direct core debris attack of the primary containment steel shell by  erecting a barrier between the seal table and the containment shell.
89 Add ribbing to the containment shell This SAMA would reduce the risk of buckling of containment under reverse pressure loading.
Improvements Related to Enhanced AC/DC Reliability/Availability 90 Proceduralize alignment of spare diesel to shutdown board after LOOP and failure of the diesel normally supplying it.
SAMA would reduce the SBO frequency.
91 Provide an additional DG.
SAMA would increase the reliability and availability of onsite emergency AC power sources.
92 Provide additional DC battery capacity.
SAMA would ensure longer battery capability during an SBO, reducing the frequency of long-term SBO sequences.
93 Use fuel cells instead of lead-acid batteries.
SAMA would extend DC power availability in an SBO.
94 Procedure to cross-tie high-pressure core spray diesel.
SAMA would improve core injection availability by providing a more reliable power supply for the high-pressure core spray pumps.
95 Improve 4.16-kV bus cross-tie ability.
SAMA would improve AC power reliability.
96 Incorporate an alternate battery charging capability.
SAMA would improve DC power reliability by either cross-tying the AC busses, or installing a portable diesel-driven battery charger.
97 Increase/improve DC bus load shedding.
SAMA would extend battery life in an SBO event.
98 Replace existing batteries with more reliable ones.
SAMA would improve DC power reliability and thus increase available SBO recovery time.
99 Mod for DC Bus A reliability.
SAMA would increase the reliability of AC power and injection capability. Loss of DC Bus A causes a loss of main condenser, prevents transfer from the main transformer to OSP, and defeats one half of the low vessel pressure permissive for low pressure coolant injection (LPCI)/CS injection valves. 100 Create AC power cross-tie capability with other unit.
SAMA would improve AC power reliability.
101 Create a cross-tie for diesel fuel oil.
SAMA would increase diesel fuel oil supply and thus DG, reliability.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-378 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 102 Develop procedures to repair or replace failed 4-kV breakers.
SAMA would offer a recovery path from a failure of the breakers that perform transfer of 4.16-kV non-emergency busses from unit station service transformers, leading to loss of emergency AC power.
103 Emphasize steps in recovery of OSP after an SBO.
SAMA would reduce HEP during OSP recovery.
104 Develop a severe weather conditions procedure.
For plants that do not already have one, this SAMA would reduce the CDF for external weather-related events.
105 Develop procedures for replenishing diesel fuel oil.
SAMA would allow for long-term diesel operation.
106 Install gas turbine generator.
SAMA would improve onsite AC power reliability by providing a redundant and diverse emergency power system.
107 Create a backup source for diesel cooling.  (Not from existing system) This SAMA would provide a redundant and diverse source of cooling for the DGs, which would contribute to enhanced diesel reliability.
108 Use FPS as a backup source for diesel cooling.
This SAMA would provide a redundant and diverse source of cooling for the DGs, which would contribute to enhanced diesel reliability.
109 Provide a connection to an alternate source of OSP.
SAMA would reduce the probability of a LOOP event.
110 Bury OSP lines.
SAMA could improve OSP reliability, particularly during severe weather.
111 Replace anchor bolts on DG oil cooler.
Millstone Nuclear Power Station found a high seismic SBO risk due to failure of the diesel oil cooler anchor bolts. For plants with a similar problem, this would reduce seismic risk. Note that these were Fairbanks Morse DGs.
112 Change undervoltage (UV), AFW actuation signal (AFAS) block and high pressurizer pressure actuation signals to 3-out-of-4, instead of 2-out-of-4 logic.
SAMA would reduce risk of 2/4  inverter failure.
113 Provide DC power to the 120/240-V vital AC system from the Class 1E station service battery system instead of its own battery. SAMA would increase the reliability of the 120-VAC Bus.
114 Bypass DG Trips SAMA would allow D/Gs to operate for longer.
115 2.i. 16 hour SBO Injection SAMA includes improved capability to cope with longer SBO scenarios.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-379 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 116 Steam Driven Turbine Generator This SAMA would provide a steam driven turbine generator which uses reactor steam and exhausts to the suppression pool. If large enough, it could provide power to additional equipment.
117 Alternate Pump Power Source This SAMA would provide a small dedicated power source such as a dedicated diesel or gas turbine for the feedwater or condensate pumps, so that they do not rely on OSP.
118 Additional DG SAMA would reduce the SBO frequency.
119 Increased Electrical Divisions SAMA would provide increased reliability of AC power system to reduce core damage and release frequencies.
120 Improved Uninterruptible Power Supplies SAMA would provide increased reliability of power supplies supporting front-line equipment, thus reducing core damage and release frequencies.
121 AC Bus Cross-Ties SAMA would provide increased reliability of AC power system to reduce core damage and release frequencies.
122 Gas Turbine SAMA would improve onsite AC power reliability by providing a redundant and diverse emergency power system.
123 Dedicated RHR (bunkered) Power Supply SAMA would provide RHR with more reliable AC power.
124 Dedicated DC Power Supply This SAMA addresses the use of a diverse DC power system such as an additional battery or fuel cell for the purpose of providing motive power to certain components (e.g., RCIC).
125 Additional Batteries/Divisions This SAMA addresses the use of a diverse DC power system such as an additional battery or fuel cell for the purpose of providing motive power to certain components (e.g., RCIC).
126 Fuel Cells SAMA would extend DC power availability in an SBO.
127 DC Cross-ties This SAMA would improve DC power reliability.
128 Extended SBO Provisions SAMA would provide reduction in SBO sequence frequencies.
129 Add an automatic bus transfer feature to allow the automatic transfer of the 120V vital AC bus from the on-line unit to the standby unit Plants are typically sensitive to the loss of one or more 120V vital AC buses. Manual transfers to alternate power supplies could be enhanced to transfer automatically.
Improvements in Identifying and Mitigating Containment Bypass
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-380 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 130 Install a redundant spray system to depressurize the primary system during a steam generator tube rupture (SGTR).
SAMA would enhance depressurization during a SGTR.
131 Improve SGTR coping abilities.
SAMA would improve instrumentation to detect SGTR, or additional system to scrub fission product releases.
132 Add other SGTR coping abilities.
SAMA would decrease the consequences of an SGTR.
133 Increase secondary side pressure capacity such that an SGTR would not cause the relief valves to lift.
SAMA would eliminate direct release pathway for SGTR sequences.
134 Replace steam generators (SG) with a new design.
SAMA would lower the frequency of an SGTR.
135 Revise emergency operating procedures to direct that a faulted SG be isolated.
SAMA would reduce the consequences of an SGTR.
136 Direct SG flooding after a SGTR, prior to core damage.
SAMA would provide for improved scrubbing of SGTR releases.
137 Implement a maintenance practice that inspects 100% of the tubes in a SG.
SAMA would reduce the potential for an SGTR.
138 Locate RHR inside of containment.
SAMA would prevent intersystem LOCA (ISLOCA) out the RHR pathway.
139 Install additional instrumentation for ISLOCAs.
SAMA would decrease ISLOCA frequency by installing pressure of leak monitoring instruments in between the first two pressure isolation valves on low-pressure inject lines, RHR suction lines, and HPSI lines.
140 Increase frequency for valve leak testing.
SAMA could reduce ISLOCA frequency.
141 Improve operator training on ISLOCA coping.
SAMA would decrease ISLOCA effects.
142 Install relief valves in the CC System.
SAMA would relieve pressure buildup from an RCP thermal barrier tube rupture, preventing an ISLOCA.
143 Provide leak testing of valves in ISLOCA paths.
SAMA would help reduce ISLOCA frequency. At Kewaunee Nuclear Power Plant, four MOVs isolating RHR from the RCS were not leak tested.
144 Revise EOPs to improve ISLOCA identification.
SAMA would ensure LOCA outside containment could be identified as such. Salem Nuclear Power Plant had a scenario where an RHR ISLOCA could direct initial leakage back to the pressurizer relief tank, giving indication that the LOCA was inside containment.
145 Ensure all ISLOCA releases are scrubbed.
SAMA would scrub all ISLOCA releases. One example is to plug drains in the break area so that the break point would be covered with water.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-381 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 146 Add redundant and diverse limit switches to each containment isolation valve.
SAMA could reduce the frequency of containment isolation failure and ISLOCAs through enhanced isolation valve position indication.
147 Early detection and mitigation of ISLOCA SAMA would limit the effects of ISLOCA accidents by early detection and isolation 148 Improved MSIV Design This SAMA would improve isolation reliability and reduce spurious actuations that could be initiating events.
149 Proceduralize use of pressurizer vent valves during steam generator tube rupture (SGTR) sequences Some plants may have procedures to direct the use of pressurizer sprays to reduce RCS pressure after an SGTR. Use of the vent valves would provide a back-up method.
150 Implement a maintenance practice that inspects 100% of the tubes in an SG This SAMA would reduce the potential for a tube rupture.
151 Locate RHR inside of containment This SAMA would prevent ISLOCA out the RHR pathway.
152 Install self-actuating containment isolation valves For plants that do not have this, it would reduce the frequency of isolation failure. Improvements in Reducing Internal Flooding Frequency 153 Modify swing direction of doors separating turbine building basement from areas containing safeguards equipment.
SAMA would prevent flood propagation, for a plant where internal flooding from turbine building to safeguards areas is a concern.
154 Improve inspection of rubber expansion joints on main condenser.
SAMA would reduce the frequency of internal flooding, for a plant where internal flooding due to a failure of circulating water system expansion joints is a concern.
155 Implement internal flood prevention and mitigation enhancements.
This SAMA would reduce the consequences of internal flooding.
156 Implement internal flooding improvements such as those implemented at Fort Calhoun.
This SAMA would reduce flooding risk by preventing or mitigating rupture in the RCP seal cooler of the component cooling system an ISLOCA in a shutdown cooling line, an AFW flood involving the need to remove a watertight door.
157 Shield electrical equipment from potential water spray SAMA would decrease risk associated with seismically induced internal flooding 158 Reduction in Reactor Building Flooding This SAMA reduces the Reactor Building Flood Scenarios contribution to core damage and release.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-382 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement Improvements Related to Feedwater/Feed and Bleed Reliability/Availability 159 Install a digital feedwater upgrade.
This SAMA would reduce the chance of a loss of main feedwater following a plant trip.
160 Perform surveillances on manual valves used for backup AFW pump suction.
This SAMA would improve success probability for providing alternative water supply to the AFW pumps.
161 Install manual isolation valves around AFW turbine-driven steam admission valves.
This SAMA would reduce the dual turbine-driven AFW pump maintenance unavailability.
162 Install accumulators for turbine-driven AFW pump flow control valves (CVs).
This SAMA would provide control air accumulators for the turbine-driven AFW flow CVs, the motor-driven AFW pressure CVs and SG power-operated relief valves (PORVs). This would eliminate the need for local manual action to align nitrogen bottles for control air during a LOOP.
163 Install separate accumulators for the AFW cross-connect and block valves This SAMA would enhance the operator's ability to operate the AFW cross-connect and block valves following loss of air support.
164 Install a new CST Either replace the existing tank with a larger one, or install a back-up tank.
165 Provide cooling of the steam-driven AFW pump in an SBO
 
event This SAMA would improve success probability in an SBO by: (1) using the FP system to cool the pump, or (2) making the pump self cooled.
166 Proceduralize local manual operation of AFW when control power is lost.
This SAMA would lengthen AFW availability in an SBO. Also provides a success path should AFW control power be lost in non-SBO sequences.
167 Provide portable generators to be hooked into the turbine driven AFW, after battery depletion.
This SAMA would extend AFW availability in an SBO (assuming the turbine driven AFW requires DC power) 168 Add a motor train of AFW to the Steam trains For PWRs that do not have any motor trains of AFW, this would increase reliability in non-SBO sequences.
169 Create ability for emergency connections of existing or alternate water sources to feedwater/condensate This SAMA would be a back-up water supply for the feedwater/condensate systems. 170 Use FP system as a back-up for SG inventory This SAMA would create a back-up to main and AFW for SG water supply.
171 Procure a portable diesel pump for isolation condenser make-up This SAMA would provide a back-up to the city water supply and diesel FP system pump for isolation condenser make-up.
172 Install an independent DG for the CST make-up pumps This SAMA would allow continued inventory make-up to the CST during an
 
SBO.
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-383 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 173 Change failure position of condenser make-up valve This SAMA would allow greater inventory for the AFW pumps by preventing CST flow diversion to the condenser if the condenser make-up valve fails open on loss of air or power.
174 Create passive secondary side coolers.
This SAMA would reduce CDF from the loss of Feedwater by providing a passive heat removal loop with a condenser and heat sink.
175 Replace current PORVs with larger ones such that only one is required for successful feed and bleed.
This SAMA would reduce the dependencies required for successful feed and bleed.
176 Install motor-driven feedwater pump.
SAMA would increase the availability of injection subsequent to MSIV closure. 177 Use Main feedwater pumps for a Loss of Heat Sink Event This SAMA involves a procedural change that would allow for a faster response to loss of the secondary heat sink. Use of only the feedwater booster pumps for injection to the SGs requires depressurization to about 350 psig; before the time this pressure is reached, conditions would be met for initiating feed and bleed. Using the available turbine driven feedwater pumps to inject water into the SGs at a high pressure rather than using the feedwater booster alone allows injection without the time consuming depressurization.
Improvements in Core Cooling Systems 178 Provide the capability for diesel driven, low pressure vessel make-up This SAMA would provide an extra water source in sequences in which the reactor is depressurized and all other injection is unavailable (e.g., FP system) 179 Provide an additional HPSI pump with an independent diesel This SAMA would reduce the frequency of core melt from small LOCA and SBO sequences 180 Install an independent AC HPSI system This SAMA would allow make-up and feed and bleed capabilities during an SBO. 181 Create the ability to manually align ECCS recirculation This SAMA would provide a back-up should automatic or remote operation fail. 182 Implement an RWT make-up procedure This SAMA would decrease CDF from ISLOCA scenarios, some smaller break LOCA scenarios, and SGTR.
183 Stop LPSI pumps earlier in medium or large LOCAs.
This SAMA would provide more time to perform recirculation swap over.
184 Emphasize timely swap over in operator training.
This SAMA would reduce HEP of recirculation failure.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-384 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 185 Upgrade Chemical and Volume Control System to mitigate small LOCAs.
For a plant like the AP600 where the Chemical and Volume Control System cannot mitigate a Small LOCA, an upgrade would decrease the Small LOCA CDF contribution.
186 Install an active HPSI system.
For a plant like the AP600 where an active HPSI system does not exist, this SAMA would add redundancy in HPSI.
187 Change "in-containment" RWT suction from 4 check valves to 2 check and 2 air operated valves.
This SAMA would remove common mode failure of all four injection paths.
188 Replace 2 of the 4 safety injection (SI) pumps with diesel-powered pumps.
This SAMA would reduce the SI system CCF probability. This SAMA was intended for the System 80+, which has four trains of SI.
189 Align low pressure core injection or core spray to the CST on loss of suppression pool cooling.
This SAMA would help to ensure low pressure ECCS can be maintained in loss of suppression pool cooling scenarios.
190 Raise high pressure core injection/RCIC backpressure trip setpoints This SAMA would ensure high pressure core injection/RCIC availability when high suppression pool temperatures exist.
191 Improve the reliability of the ADS.
This SAMA would reduce the frequency of high pressure core damage sequences.
192 Disallow automatic vessel depressurization in non-ATWS scenarios This SAMA would improve operator control of the plant.
193 Create automatic swap over to recirculation on RWT depletion This SAMA would reduce the human error contribution from recirculation failure. 194 Proceduralize intermittent operation of high pressure coolant injection (HPCI).
SAMA would allow for extended duration of HPCI availability.
195 Increase available NPSH for injection pumps.
SAMA increases the probability that these pumps will be available to inject coolant into the vessel by increasing the available NPSH for the injection
 
pumps. 196 Modify Reactor Water Cleanup (RWCU) for use as a decay heat removal system and proceduralize use.
SAMA would provide an additional source of decay heat removal.
197 Control Rod Drive (CRD) Injection SAMA would supply an additional method of level restoration by using a non-safety system.
198 Condensate Pumps for Injection SAMA to provide an additional option for coolant injection when other systems are unavailable or inadequate
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-385 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 199 Align EDG to CRD for Injection SAMA to provide power to an additional injection source during loss of power events 200 Re-open MSIVs SAMA to regain the main condenser as a heat sink by re-opening the MSIVs. 201 Bypass RCIC Turbine Exhaust Pressure Trip SAMA would allow RCIC to operate longer.
202 Passive High Pressure System SAMA will improve prevention of core melt sequences by providing additional high pressure capability to remove decay heat through an isolation condenser type system 203 Suppression Pool Jockey Pump SAMA will improve prevention of core melt sequences by providing a small makeup pump to provide low pressure decay heat removal from the RPV using the suppression pool as a source of water.
204 Improved High Pressure Systems SAMA will improve prevention of core melt sequences by improving reliability of high pressure capability to remove decay heat.
205 Additional Active High Pressure System SAMA will improve reliability of high pressure decay heat removal by adding an additional system.
206 Improved Low Pressure System (Firepump)
SAMA would provide FPS pump(s) for use in low pressure scenarios.
207 CUW Decay Heat Removal This SAMA provides a means for Alternate Decay Heat Removal.
208 High Flow Suppression Pool Cooling SAMA would improve suppression pool cooling.
209 Diverse Injection System SAMA will improve prevention of core melt sequences by providing additional injection capabilities.
210 Alternate Charging Pump Cooling This SAMA will improve the high pressure core flooding capabilities by providing the SI pumps with alternate gear and oil cooling sources. Given a total loss of Chilled Water, abnormal operating procedures would direct alignment of preferred Demineralized Water or the Fire System to the Chilled Water System to provide cooling to the SI pumps' gear and oil box (and the other normal loads).
Instrument Air/Gas Improvements 211 Modify EOPs for ability to align diesel power to more air compressors.
For plants that do not have diesel power to all normal and back-up air compressors, this change would increase the reliability of IA after a LOOP.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-386 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 212 Replace old air compressors with more reliable ones This SAMA would improve reliability and increase availability of the IA compressors.
213 Install nitrogen bottles as a back-up gas supply for SRVs.
This SAMA would extend operation of SRVs during an SBO and loss of air events (BWRs).
214 Allow cross connection of uninterruptible compressed air supply to opposite unit.
SAMA would increase the ability to vent containment using the hardened vent. ATWS Mitigation 215 Install MG set trip breakers in CR This SAMA would provide trip breakers for the MG sets in the CR. In some plants, MG set breaker trip requires action to be taken outside of the CR. Adding control capability to the CR would reduce the trip failure probability in sequences where immediate action is required (e.g., ATWS).
216 Add capability to remove power from the bus powering the control rods This SAMA would decrease the time to insert the control rods if the reactor trip breakers fail (during a loss of feedwater ATWS which has a rapid pressure excursion) 217 Create cross-connect ability for standby liquid control trains This SAMA would improve reliability for boron injection during an ATWS
 
event. 218 Create an alternate boron injection capability (back-up to standby liquid control)
This SAMA would improve reliability for boron injection during an ATWS
 
event. 219 Remove or allow override of low pressure core injection during an ATWS On failure on high pressure core injection and condensate, some plants direct reactor depressurization followed by 5 minutes of low pressure core injection. This SAMA would allow control of low pressure core injection immediately.
220 Install a system of relief valves that prevents any equipment damage from a pressure spike during an ATWS This SAMA would improve equipment availability after an ATWS.
221 Create a boron injection system to back up the mechanical control rods.
This SAMA would provide a redundant means to shut down the reactor.
222 Provide an additional instrument system for ATWS mitigation (e.g., ATWS mitigation scram actuation circuitry).
This SAMA would improve instrument and control redundancy and reduce the ATWS frequency.
223 Increase the SRV reseat reliability.
SAMA addresses the risk associated with dilution of boron caused by the failure of the SRVs to reseat after standby liquid control (SLC) injection.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-387 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 224 Use CRD for alternate boron injection.
SAMA provides an additional system to address ATWS with SLC failure or unavailability.
225 Bypass MSIV isolation in Turbine Trip ATWS scenarios SAMA will afford operators more time to perform actions. The discharge of a substantial fraction of steam to 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 226 Enhance operator actions during ATWS SAMA will reduce human error probabilities during ATWS 227 Guard against SLC dilution SAMA to control vessel injection to prevent boron loss or dilution following SLC injection.
228 ATWS Sized Vent This SAMA would provide the ability to remove reactor heat from ATWS events. 229 Improved ATWS Capability This SAMA includes items which reduce the contribution of ATWS to core damage and release frequencies.
Other Improvements 230 Provide capability for remote operation of secondary side relief valves in an SBO Manual operation of these valves is required in an SBO scenario. High area temperatures may be encountered in this case (no ventilation to main steam areas), and remote operation could improve success probability.
231 Create/enhance RCS depressurization ability With either a new depressurization system, or with existing PORVs, head vents, and secondary side valve, RCS depressurization would allow earlier low pressure ECCS injection. Even if core damage occurs, low RCS pressure would alleviate some concerns about HPME.
232 Make procedural changes only for the RCS depressurization option This SAMA would reduce RCS pressure without the cost of a new system 233 Defeat 100% load rejection capability.
This SAMA would eliminate the possibility of a stuck open PORV after a LOOP, since PORV opening would not be needed.
234 Change CRD flow CV failure position Change failure position to the "fail-safest" position.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-388 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 235 Install secondary side guard pipes up to the MSIVs This SAMA would prevent secondary side depressurization should a steam line break occur upstream of the MSIVs. This SAMA would also guard against or prevent consequential multiple SGTR following a Main Steam Line Break event.
236 Install digital large break LOCA protection Upgrade plant instrumentation and logic to improve the capability to identify symptoms/precursors of a large break LOCA (leak before break).
237 Increase seismic capacity of the plant to a high confidence, low pressure failure of twice the Safe Shutdown Earthquake.
This SAMA would reduce seismically -induced CDF.
238 Enhance the reliability of the demineralized water (DW) make-up system through the addition of diesel-backed power to one or both of the DW make-up pumps.
Inventory loss due to normal leakage can result in the failure of the CC and the SRW systems. Loss of CC could challenge the RCP seals. Loss of SRW results in the loss of three EDGs and the containment air coolers (CACs). 239 Increase the reliability of SRVs by adding signals to open them automatically.
SAMA reduces the probability of a certain type of medium break LOCA. Hatch evaluated medium LOCA initiated by an MSIV closure transient with a failure of SRVs to open. Reducing the likelihood of the failure for SRVs to open, subsequently reduces the occurrence of this medium LOCA.
240 Reduce DC dependency between high-pressure injection system and ADS.
SAMA would ensure containment depressurization and high-pressure injection upon a DC failure.
241 Increase seismic ruggedness of plant components.
SAMA would increase the availability of necessary plant equipment during and after seismic events.
242 Enhance RPV depressurization capability SAMA would decrease the likelihood of core damage in loss of HPCI scenarios 243 Enhance RPV depressurization procedures SAMA would decrease the likelihood of core damage in loss of HPCI scenarios 244 Replace mercury switches on FPSs SAMA would decrease probability of spurious fire suppression system actuation given a seismic event+D114 245 Provide additional restraints for CO 2 tanks SAMA would increase availability of FP given a seismic event.
246 Enhance control of transient combustibles SAMA would minimize risk associated with important fire areas.
247 Enhance fire brigade awareness SAMA would minimize risk associated with important fire areas.
248 Upgrade fire compartment barriers SAMA would minimize risk associated with important fire areas.
 
Environmental Report Appendix E SAMA ANALYSIS Three Mile Island Nuclear Station Unit 1 Page E-389 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 249 Enhance procedures to allow specific operator actions SAMA would minimize risk associated with important fire areas.
250 Develop procedures for transportation and nearby facility accidents SAMA would minimize risk associated with transportation and nearby facility
 
accidents.
251 Enhance procedures to mitigate Large LOCA SAMA would minimize risk associated with Large LOCA 252 Computer Aided Instrumentation SAMA will improve prevention of core melt sequences by making operator actions more reliable.
253 Improved Maintenance Procedures/Manuals SAMA will improve prevention of core melt sequences by increasing reliability of important equipment 254 Improved Accident Management Instrumentation SAMA will improve prevention of core melt sequences by making operator actions more reliable.
255 Remote Shutdown Station This SAMA would provide the capability to control the reactor in the event that evacuation of the MCR is required.
256 Security System Improvements in the site's security system would decrease the potential for successful sabotage.
257 Improved Depressurization SAMA will improve depressurization system to allow more reliable access to low pressure systems.
258 Safety Related CST SAMA will improve availability of CST following a Seismic event 259 Passive Overpressure Relief This SAMA would prevent vessel overpressurization.
260 Improved Operating Response Improved operator reliability would improve accident mitigation and prevention.
261 Operation Experience Feedback This SAMA would identify areas requiring increased attention in plant operation through review of equipment performance.
262 Improved SRV Design This SAMA would improve SRV reliability, thus increasing the likelihood that sequences could be mitigated using low pressure heat removal.
263 Increased Seismic Margins This SAMA would reduce the risk of core damage and release during seismic events.
264 System Simplification This SAMA is intended to address system simplification by the elimination of unnecessary interlocks, automatic initiation of manual actions or redundancy as a means to reduce overall plant risk.
 
Environmental Report Appendix E  SAMA ANALYSIS Page E-390 Three Mile Island Nuclear Station Unit 1 License Renewal Application TABLE A-1 SELECTED PREVIOUS INDUSTRY SAMAs SAMA ID number SAMA title Result of potential enhancement 265 Train operations crew for response to inadvertent actuation signals This SAMA would improve chances of a successful response to the loss of two 120V AC buses, which may cause inadvertent signal generation.
266 Install tornado protection on gas turbine generators This SAMA would improve onsite AC power reliability.}}

Latest revision as of 18:43, 14 January 2025

License Renewal Application, Environmental Report Appendix E, Table of Contents Through Page E-390
ML080220282
Person / Time
Site: Crane Constellation icon.png
Issue date: 01/08/2008
From:
AmerGen Energy Co
To:
Office of Nuclear Reactor Regulation
References
5928-08-20001
Download: ML080220282 (402)
v • d • e


Text

Retrieved from "https://kanterella.com/w/index.php?title=ML080220282&oldid=5336333"