ML17348A282

Suppl 2,Rev 1 to Emergency Power Sys Enhancement Rept, Safety Analysis.

ML17348A282
Person / Time
Site:	Turkey Point
Issue date:	05/31/1990
From:	FLORIDA POWER & LIGHT CO.
To:
Shared Package
ML17348A279	List: ML17348A278 ML17348A280 ML17348A281 ML17348A282 ML17348A283
References
NUDOCS 9006120233
Download: ML17348A282 (85)

Text

0 FLORIDA POWER 5 LIGHT CPMPANY TURKEY POINT UNITS 3 AND 4 EMERGENCY POWER SYSTEM ENHANCEMENT REPORT SUPPLEMENT NO. 2 SAFETY ANALYSIS REVISION 0 MAY 1990

~, 9006120233 900604 PDR ADOCK 05000250 PDC F

TABLE OF CONTENTS SECTION TITLE PAGE

2.0 DESCRIPTION

OF ENHANCED EMERGENCY POWER SYSTEM EPS 2-'1 2.1 OVERVIEW 2-1 2.2 ELECTRICAL/INSTRUMENTATION 8L CONTROL MODIFICATIONS 2-3 2.3 MECHANICAL AND STRUCTURAL ADDITIONS 2-4 2.4 RESPONSES TO NRC REQUESTS FOR ADDITIONAL INFORMATION 3.0 EDG LOADING UNDER DESIGN BASIS ACCIDENT CONDITIONS 3-1 3.1 EXISTING EPS 3-1 3.2 ENHANCED EPS WITH NORMAL (TRAIN 8) ALIGNMENT 3-1 3.3 ENHANCED EPS WITH ALTERNATE (TRAIN A) ALIGNMENT 3-3 3.4 ENHANCED EPS RESULTS 3-4 3.5 LONG TERM EDG LOADING 3-5

3.6 CONCLUSION

S 3-9 4.0 SE UENCER ENHANCEMENTS 4-1 4.1 EXISTING LOAD BLOCKS 4-1 4.2 ENHANCED LOAD BLOCKS 4-3 4.3 SEQUENCING LOADS ONTO BUSES WITH OFFSITE POWER AVAILABLE 4-4 ENHANCED SEOUENCER.DESIGN TO MEET SINGLE FAILURE CRITERIA 5.0 FAILURE MODES AND EFFECTS ANALYSES FMEAs 5-1 5.1 FMEA OF THE ONSITE AC EMERGENCY POWER SYSTEM 5-1 5.2 FMEA OF THE 125 V DC ENHANCEMENTS 5-2

TABLE OF CONTENTS (Continued)

SECTION TITLE PAGE 6.0 PROBABILISTIC EVALUATIONS 6-1

6. 1 PURPOSE AND SCOPE 6-1

6. 2 QUANTITATIVE EVALUATION METHODOLOGY 6-1 6.3 'RESULTS 6-4

7.0 CONCLUSION

S 7-1

8.0 REFERENCES

, 8-1 tj gr'

LIST OF TABLES TABLE TITLE 3-1 Short-Term EDG kW Loads for Unit 3 LOOP plus LOCA, Two-Unit Operation with Swing Loads Aligned to B Train 3-2 Short-Term EDG kW Loads for Unit 3 LOOP plus LOCA, Two-Unit Operation with Swing Loads Aligned to A Train 3-3 Long-Term EDG kW Loads for Unit 3 LOOP plus LOCA, Two-Unit Operation with Swing Loads Aligned to B Train 3-4 Long-Term EDG kW Loads for Unit 3 LOOP plus LOCA, Two-Unit Operation with Swing Loads Aligned to A Train 6-1 Definition of 4KV Bus States 6-2 Definition of Plant States LIST OF FIGURES FIGURE TITLE AC One-Line Diagram DC One-Line Diagram

-111-

I' t

C

1. 0 INTRODUCTION Florida Power and Light Company (FPL) is installing two new emergency diesel generators (EDGs) and associated electrical and mechanical equipment at the Turkey Point. Nuclear site, as documented in the Emergency Power System (EPS) Enhancement Report enclosed with FPL letter L-88-269 dated June 23, 1988. Supplement I to the EPS Enhancement Report was submitted via FPL letter L-89-124 dated April 3, 1989 and provided information regarding the testing to be performed on the various components and systems during turnover and startup, during preoperational testing, and prior to returning the enhanced EPS to service.

This Supplement 2 to the EPS Enhancement Report provides FPL's Safety Analysis for the enhanced EPS configuration. As discussed herein, the enhanced EPS configuration provides an improved response to the existing FSAR limiting Design Basis Accident (DBA) by providing enhanced equipment availability on the accident unit with increased EDG loading margin. From an EDG loading standpoint, the existing FSAR limiting DBA is defined as both Units at 100% power, a LOOP on both Units, a LOCA on one Unit, and a single active failure of an EDG.

In the existing design, one EDG must provide the capacity to-mitigate the LOOP/LOCA in one Unit and attain safe (hot) shutdown in the other Unit. The enhanced design alleviates this dual-Unit reliance on one EDG in the following ways:

I. As indicated in Reportable Event 85-42 (Revision I), documented in FPL's letter to the NRC L-86-256, a potential existed prior to November 1985 for exceeding the limits for EDG loading during the DBA. The resolution of this issue provides a transient, and .short-term continuous, load limit philosophy of 2950/2850 kW, loaded on one existing EDG for. the Design Basis Accident described above. Operator actions and load management are required for the existing design to ensure that the EDG is not overloaded. The enhanced configuration improves the plant response during this DBA since:

a) The auto-connect loads on each Unit's EDG(s) for a LOOP, or LOOP and LOCA, are within the continuous rating of each EDG (2500 kW for the existing EDGs, 2874 kW for the new EDGs); and b) The load limit philosophy for the new design stays below the continuous rating of each EDG during a DBA even with auto-connect and concurrent manual loads considered.

x

/4'I t

% l>

t>

Wa f

NP Vl

2. The enhanced EPS configuration ensures that a minimum of three EOGs are operable in response to the OBA, versus only one EOG in the existing configuration. Since in the enhanced design each EOG powers one 4. 16 kV bus, and its associated equipment, three 4.16 kV buses and equipment are available for the OBA (as discussed below), versus only two 4. 16 kV buses and the minimum ESF in the existing design (due to EDG loading limits). The enhanced EPS thus provides additional equipment to either mitigate the accident or to shut down the non-accident Unit (depending on which Unit the failure of an EOG is postulated) as follows:

a) If the EDG failure is on the non-accident Unit, the operators have more flexibility to manage the accident using the two EDGs and associated equipment available on the accident Unit, plus an additional High-Head Safety Injection (HHSI) Pump available from the non-accident Unit's OPERABLE EDG (assuming, that an OPERABLE HHSI pump was aligned to the remaining EDG). The operators on the accident Unit can utilize at least two HHSI pumps, both CS pumps, two RHR pumps, two CCW pumps, two ICW pumps, all three ECCs, all three ECFs, all three CRAC units, plus plant investment loads (e.g., the turbine-generator loads) as desired. In this scenario, none of the EOGs assumed available exceed their continuous rating. This allows more ESF equipment available than was assumed in the FSAR.

b) If the EDG failure is on the accident Unit, the operators have more than the minimum equipment available .than was assumed in the FSAR, plus an additional HHSI Pump. That is, the operators on the accident Unit can utilize at least two HHSI pumps, two ECCs, two ECFs, all three CRACs, plus plant investment loads. The FSAR OBA analysis remains the bounding accident due to the minimum equipment assumed available.

This Safety Analysis was developed in accordance with the requirements of the NRC's Standard Review Plan (NUREG-0800, Section 8.3.1), which states that for shared electrical configurations, sufficient analyses shall be provided which substantiate the adequacy of the design to withstand the consequences of electrical faults and failures in one Unit with respect to the other Unit(s).

Therefore in accordance with Standard Review Plan requirements, this Safety Analysis demonstrates the following:

1. The enhanced EPS configuration improves overall plant safety by essentially doubling the available capacity of Turkey Point on-site EPS. Under design basis conditions (which include single 1-2

J"~

kf i1 T

1+4 r

failure considerations), EDG loading is maintained within each EDG's continuous rating.

2. Required Engineered Safety Feature loads and desired plant investment loads are accommodated with the enhanced EPS configuration, while retaining the shared systems as originally designed.

In addition to the above, this Safety Analysis. provides a discussion of the quantitative probabilistic evaluations performed for both the existing and the enhanced EPS 4. 16 kV bus cpnfigurations. These evaluations (considering the,AC power recovery capability of both

. the existing system capability and the inter-Unit crosstie provided by the enhanced design, and considering a conservative probability of operator error for either design), show that the 4. 16 kV bus state failure frequency is reduced with the enhanced design. Hence overall plant safety as measured by the availability of emergency power to the plant's safety buses is improved under the enhanced EPS configuration.

Evaluations of the enhanced EPS will verify that the current design basis accident analyses as presented in the FSAR are not adversely impacted and remain valid under the enhanced EPS configuration.

This Safety Analysis is divided into eight sections. Following this introduction, Section 2.0 presents an overview of the enhanced EPS design and identifies any significant changes which have occurred since issuance of the June 23, 1988 EPS Enhancement Report. Section 3.0 provides analyses to show that the availability of power to required Engineered Safety Feature loads and desired plant investment loads is assured under the enhanced EPS configuration without exceeding the continuous rating of any EDG. Section 4.0 describes the enhanced sequencer design. Section 5.0 provides the results of Failure Modes and Effects Analyses (FHEAs) for postulated DBA single failures to*show that under design basis conditions, the enhanced EPS will accomplish its required safety function. Section 6.0 provides the results of a quantitative probabilistic evaluation which compares the enhanced EPS to the existing EPS with respect to their ability to successfully provide power to the 4.16 kV buses.

Section 7.0 then provides a summary and conclusion. References used in the Safety Analyses are listed in Section 8.0.

1-3

~r t~l t'l OJ Tl-V

• 'p F

lf l

2.0 DESCRIPTION

OF ENHANCED EMERGENCY POWER SYSTEN 2.1 OVERVIEW The enhanced EPS includes the installation of two new EDGs with all support systems (fuel oil, starting air, ventilation, etc.), a new emergency diesel generator building, diesel oil storage tanks and transfer pumps in an associated building, new 4. 16 kV switchgears, new 480V load centers, new 480V motor control centers, new 125V DC transfer/distribution panels, new sequencers, breakers, battery chargers, etc., plus lighting distribution panels, transformers, cabling and numerous components necessary for modifying the existing equipment.

The design of the enhanced EPS also meets the Station Blackout Rule, 10CFR50.63, by adding an intertie between each Unit's new "D" (3D and 4D) 4. 16 kV switchgear. This feature provides an alternate AC power supply to the blackout Unit through the use of an operable EDG on the non-blackout unit. This Turkey Point Plant conformance to 10CFR50.63 is the subject of a separate submittal to NRC. (Refer to FPL letter L-89-144, dated April 17, 1989).

See Figures 1 and 2 for a one-line electrical diagram of the AC and DC systems, respectively. Note these figures are essentially the same as those included in the June 1988 EPS Enhancement Report, but have been redrawn for clarity and to reflect several changes pertaining to the proposed AC and DC system modifications.

The new seismic Category I diesel building is located northeast of the Unit 3 contai,nment. The building is two stories high with the diesel generators located on the lower elevation and the auxiliaries such as air start skids, control panels, motor control centers, distribution panels, etc., located on the upper level. Also located on the upper level are the two new 4. 16 kV swing buses, one for each Unit.

As part of the EPS Enhancement Project, existing EDG ¹3 (EDG "A"),

presently supplying power to the A system of both Units, is reassigned to the Unit 3A power system, and relabeled EDG 3A.

Similar ly, existing EDG ¹4 (EDG "B") is relabeled EDG 3B and assigned to supply power to the Unit 3B power system. Thus, the two existing EDGs are aligned as the emergency AC power supplies for Unit 3 and certain common or shared systems.

The two new EDGs are aligned as the emergency AC power supplies for Unit 4 and certain common or shared systems. EDG 4A supplies the Unit 4A power system and EDG 4B supplies the Unit 4B power system.

2-1

e l~

"4 fW 4

C, t%

III 4

The four existing sequencers are replaced with four new solid-state sequencers.

The new swing 4. 16 kV Switchgear 30 supplies power to the Intake Cooling Water ( ICW) Pump 3C, Component Cooling Water (CCW) Pump 3C and the intertie breaker for station blackout; likewise, the new swing Switchgear 40 supplies power to ICW Pump 4C, CCW Pump 4C and the intertie breaker for station blackout. Refer to Figure l.

During normal operation, the swing 4. 16 kV switchgear power supply breakers can be manually aligned to either the A or B switchgear.

Each Unit has a new 480V load center swing bus both located in the Auxiliary Building. The new swing 480V Load Center 3H supplies power to HCC 30 and to Charging Pump (CP) 3C; likewise, the new swing Load Center 4H supplies power to MCC 40 and to Charging Pump 4C. Refer to Figure l. Each swing load center can be aligned to Train A or Train B of its associated Unit. However, during power operation, the alignment of these swing load centers will normally be to the B Train of each Unit. The EDG loading is acceptable with either Train alignment. For each 480V swing load center, if the bus to which it is aligned loses power, automatic circuit breaker operation connects the bus to the alternate power source.

A new. Hotor Control Center (HCC) 3K is added to supply EDG 3B auxiliaries (presently supplied from Unit 4 HCC 4B). Existing HCC 0, (which presently supplies power to Unit 3 and 4 third service loads and plant common loads), is relabeled MCC 30, and supplies power to Unit 3 loads and existing plant common loads. New HCC 40 is added to supply vital loads associated with Unit 4 that are presently fed from HCC 0. New HCCs 4J and 4K are added to power the auxiliary loads for the new EDGs. Refer to Figure 1. The existing EOG ¹3 (renamed EDG 3A) auxiliaries are powered from HCC. 3A and are not affected.

The existing HCCs 3A and 4A have Telemand transfer systems which presently allow them to be powered from either existing Train A or Train B. These existing Telemand operators will be removed as the safety loads connected to these HCCs have redundant counterparts.

With the enhanced design, HCC 3A and 4A are powered from the Train A of each respective Unit and the redundant safety loads are powered from Train B and/or the new swing LCs/HCCs.

The existing HCC 0 provides power to common, shared and third service loads. This HCC has a Telemand Transfer System which allows it to be powered from either Unit 3 (Train B) or Unit 4 (Train A).

This existing Telemand operator will be removed. The enhanced design thus eliminates the Telemand logic and replaces it with a power-seeking transfer design at the (new) Load Center H level.

Loads which previously could be powered from either Unit 3 or Unit 4 are relocated as shown on Figure I, and now can be powered from either the "A" or "B" Train of each Unit including the swing loads.

2-2

i 4

I pV hf l

The six existing battery chargers are being replaced with six new battery chargers, and are being relabeled as indicated on Figures 1 and 2. Two additional battery chargers, 3A2 and 3B2, are added as new equipment, and the two existing spare battery chargers 3S and 4S are being realigned and relabeled, whereby 3S becomes 4B2 and 4S becomes 4A2. These modifications provide the capability of two independently powered battery chargers aligned to each DC bus, as shown in Figures 1 and 2. The battery chargers 3A2 and 382 are powered from the new MCC 40, and the relabeled battery chargers 4A2 and 4B2 are powered from MCC 3D (relabeled). With two battery chargers able to be aligned to each DC bus, a single failure of a battery charger or its MCC still assures that the redundant battery charger can be made available for its associated DC bus. Refer to Figures 1 and 2; also see the discussions in Section 5.0, Failure Modes and Effects Analyses (FMEA).

As a result of these additions and modifications, Turkey Point Units 3 and 4 have a safer, more reliable system with the capability of having one train out of service without significantly affecting the other Unit. The use of swing bus arrangements, to power the third ESF loads of the two out of three ESF equipment, provides the capability to power those loads from either Train on a Unit.

Components are more available for maintenance since the new plant alignments allow items to be taken out of service with lessened Technical Specification impact on the other Unit. The increase in emergency generation capacity and the addition of switchgears, motor control centers and distribution panels allows future load growth when required and the ability to add plant investment protection loads upon completion of the modifications.

2.2 ELECTRICAL/INSTRUMENTATION AND CONTROL MODIFICATIONS The June 1988 EPS Enhancement Report (L-88-269), describes the electrical and instrumentation/controls modifications being performed under the EPS Enhancement Project. Please refer to the Emergency Power System Enhancement Report (EPSER) for information relating to the modifications being made.

In addition to the June 1988 EPSER, additional information is provided (under separate cover) by the FPL responses to NRC's RAIs (see 2.4 below), by Supplement 1: Testing, and by this Supplement 2: Safety Analysis.

Since issuance of the above documents, two additional changes have been made with respect to the planned DC system modifications. The changes have been incorporated into Figure 2 and include the following:

2-3

k

'~

h ~

r,)

1. The two new battery chargers being installed under the EPS Enhancement Project (labeled 3A2 and 382) are assigned to DC buses 3D01 and 3D23 respectively. The two existing swing chargers (relabeled 4A2 and 482) are assigned to DC buses 4D23 and 4D01 respectively.

2. The tie-line between DC buses 3D23 and 4D23 and the existing tie-line between DC buses 3D01 and 4D01 will be used with a new safety-related spare battery. With a safety-related spare battery, it will be possible to test each of the existing station batteries while its affected DC bus is aligned to the spare battery.

2.3 MECHANICAL AND STRUCTURAL ADDITIONS Details of the mechanical and structural additions are provided in the June 1988 EPSER, and in the FPL responses to NRC's RAIs (see 2.4 below). Please refer to that information as necessary.

In addition, since issuance of the above documents, the following change has occurred to the design of the new EDG building:

1. The June 1988 EPS Enhancement Report, Section 5. 1, states that missile protection for structures is provided by reinforced concrete walls, heavy steel grating, and exterior steel missile doors.. However, in lieu of the missile doors, the new EDG building configuration has been revised to include the installation of reinforced concrete labyrinths which serve to protect exterior doors from impact by postulated missiles. The use of such concrete labyrinths provides missile protection capability equivalent to that achieved through the use of the previously specified steel doors.

2.4 RESPONSES TO NRC REQUEST FOR ADDITIONAL INFORMATION Following submittal of the June 1988 EPSER, the April 1989 Testing Submittal (Supplement 1) and additional letter information from FPL, in January 1989 and in March 1990 the NRC transmitted Requests for Additional Information (RAIs) regarding the EPSER, pre-op testing, and FPL's position concerning waiver of the 300-start and load qualification testing of the new EDGs.

FPL's responses to the RAIs were forwarded under separate cover and provide additional information related to the June 1988 EPSER, the Supplement 1, and to this Supplement 2. Please refer to all of the above documents, in addition to this Supplement 2, for further information regarding the enhanced EPS.

2-4

0 3.0 EDG LOADING UNDER DESIGN BASIS ACCIDENT CONDITIONS 3.1 EXISTING EPS For the existing emergency power system, the most limiting Design Basis Accident (DBA) with respect to EDG loading is a large break LOCA (LBLOCA), coincident with a loss of offsite power (LOOP) and a single failure of one of the two EDGs to start. The highest EDG loads were shown to occur in the initial short-term phases of the accident, i.e., in the first 0-30 minutes. Both units are assumed initially at full power in a normal operating condition (i.e., no equipment is out of service or undergoing test).

Specifically, the most limiting DBA for the existing EPS is a LBLOCA on Unit 3 with failure of the 83 (or "A") EDG to start. This scenario imposes the maximum loads on the remaining EDG, which must provide power to the accident Unit loads to mitigate the accident (using the maximum amount of Engineered Safety Features available),

and must also provide power to the non-accident Unit loads to achieve Hot Standby conditions.

It should be noted that the ability of the existing EPS to meet this design basis was the subject of an April 2, 1986 NRC Confirmation of Action Letter (CAL) which limited total loads on the EDGs to no more than 2845 kW per EDG. FPL submitted a Safety Evaluation for dual-Unit operation via letter L-86-243 dated June 12, 1986 and responded to NRC's July 8 Request for Additional Information (RAI) via letter L-86-295 dated July 16. As a result, the NRC issued a Safety Evaluation (SE) on December 15, 1986. The NRC concluded that the loads for the various conditions were in conformance with Regulatory Guide 1.9, Position C.2 and that the actions identified in the CAL were completed. A transient/short-term continuous load limit philosophy of 2950/2850 kW was found acceptable in the NRC SE.

3.2 ENHANCED EPS WITH NORMAL (TRAIN B) ALIGNMENT 3.2.1 LOCA on Unit 3 The enhanced EPS design, with four EDGs each powering a 4. 16 kV bus, ensures that for the same design basis accident described above (i.e., LOOP on both Units, LOCA on one Unit, and failure of one EDG) there are 3 out of 4 EDGs available, with one 4.16 kV bus powered on one Unit, and two 4.16 kV buses powered on the other Unit. Both Units are assumed initially at full power in a normal operating condition (i.e., no equipment is out of service or undergoing test).

Only one 4.16 kV bus on each Unit is required for successful mitigation of an accident on the Unit suffering the LOOP and LOCA, and for successful safe (hot) shutdown for the Unit undergoing a LOOP.

This Subsection and the next discuss various accident scenarios and illustrate that accident mitigation and safe shutdown are accomplished without exceeding any one EDG's continuous rating.

Since provision of a swing Load Center "H" (i.e., Load Centers 3H and 4H) on each Unit allows alignment of the vital MCCs "D" (i.e.,

MCCs 3D and 4D) to either Train A or Train B of each Unit, this 3-1

II 4 p I(

I

~1 l-*

Subsection discusses the accident scenarios (dual Unit .LOOP, single Unit LOCA) with the load centers in their "normal" alignment to the B Train. Subsection 3.3 then discusses the LOOP/LOCA accident scenarios for the case where the swing load centers are assumed to be aligned to the A Train of each Unit. Demonstration that the required safe shutdown and accident mitigation loads are always available in the short-term of the accident (i.e., the first hour) is provided in this and the next Subsection and summarized in Subsection 3.4. Subsection 3.5 demonstrates that the required safe shutdown and accident mitigation loads are available in the long-term (i.e. about eight hours) after the accident. Subsection 3,6 provides overall conclusions.

As indicated in the previous Subsection, the most limiting DBA for the existing EPS is a LBLOCA on Unit 3 with failure of the existing "A" EDG to start. Table 3-1 shows the enhanced EPS EDG short-term loads for all four EDGs, for a LOCA on Unit 3. (No single failure of an EDG is shown in Table 3-1, but is discussed later). Table 3-1 identifies the continuous load rating for the existing and the new EDGs, and illustrates the additional EDG capacity which is now provided by the enhanced EPS. Some discussion of the information provided in Table 3-1, and the other Tables in this Section 3.0, is required:

The existing component kW loads were reanalyzed in 1985-1986 during the evaluation of the potential EDG loading concern, and these kW values are reflected in the updated FSAR (Reference 1). To ensure that there is no EDG loading concern with the enhanced EPS, the component kW values shown in the Section 3.0 Tables are equal to, or greater than, the component kW values presented in the updated FSAR. The kW values used in Section 3.0 are equal to, or greater than, the highest power requirements of a motor (such as a Containment Spray Pump motor, or an RHR Pump motor) assuming maximum head and flow conditions, and running singly. In addition to this conservatism, the loading tables in this Section also depict various manual loads (shown in parentheses in the tables) also added onto each EDG along with the auto-connect loads. In addition, a conservative estimate of miscellaneous loads (35 kW) was added to each EDG, and a conservative estimate of plant investment loads was added to each EDG (i.e., 175 kW on each Train A EDG, and 100 kW on each Train B EDG). Finally, the loading tables show a total EDG load as if all the loads shown are run concurrently, when in the loads are not all powered at the same time.

reality By inspection of Table 3-1, it can be seen that the loading on each EDG is within its continuous rating, even with the assumption of concurrent manual loads imposed. Therefore, with one EDG powering each 4. 16 kV bus, the enhanced design alleviates existing EDG load management scenarios by providing additional load capacity. Also by inspection of Table 3-1, it can be seen that the failure of one EDG on the accident Unit (Unit 3) does not result in the loss of the minimum required Engineered Safety Features, due to the automatic swing of LC 3H/HCC 3D if power is lost. Also see Figure 1: if 3-2

I k,

f F}

ldll

",I

power is lost to LC 30 (by failure of the 3B EDG, or some other single failure), the loads shown on HCC 30 are still available since LC 3H/NCC 3D automatically swing to the alternate power source, LC 3C. Similarly, a loss of an EDG on the LOOP Unit (Unit 4) does not result in the loss of the minimum equipment required for hot shutdown.

3.2.2 LOCA on Unit 4 For a LBLOCA on Unit 4, with offsite power not available and with the same normal alignment (8 Train) loadings assumed for Table 3-1, the EDG short-term loads were calculated and result in the following (with no single failure'assumed):

EDG 3A kW EDG 3B kW EDG 4A kW EDG 4B kW 2016 2204 1927 2210 Thus, if a LBLOCA totals again indicate, occurs on Unit even 4 instead of Unit 3, the above for the concurrent loads assumed, that each EDG kW load is still well below its continuous rating. The failure of any one EDG still results in sufficient equipment for accident mitigation on Unit 4 and Hot Standby on Unit 3.

3.3 ENHANCED EPS WITH ALTERNATE (TRAIN A) ALIGNMENT 3.3.1 LOCA on Unit 3 As discussed in Subsection 3:2.1, loss of the normal power source to a swing load center on either Unit results in automatic transfer of equipment powered from the 3H/4H Load Center(s) from the normal B Train alignment to an A Train alignment. This alignment can also be used during normal operation, and is administratively controlled and interlocked to preclude paralleling load centers. For the purpose of this analysis, this realignment of loads, and the resultant EDG kW loadings were evaluated for a LBLOCA on Unit 3, where the results are shown assuming both Units were operating with the "A" Train alignment. For this condition, Table 3-2 shows the

'DG short-term loads which could result in approximately the first hour with'out the assumption of any single failure.

By inspection of Table 3-2, it can be seen that, even with the assumption of concurrent manual loads imposed, each EDG kW load is still well below its continuous rating. Note also that, as in Table 3-1, component kW load values generally have been increased from the values presently assumed in the FSAR.

In addition, by inspection of Tables 3-1 and Table 3-2, it can be seen that the required short-term loads are still available even if EDG 3A (or EDG 38) is assumed to fail. The requirement for an additional High Head Safety Injection Pump (since one is lost by failure of a Unit 3 EDG) is met by EDG 4A, with yet another HHSI Pump available if desired from EDG 4B (or vice-versa). The loss of the other short-term components needed is accommodated by their 3 3

v~

l ~

~ 4 4

i4

'ag aPi

,Cl t pl D

"p A

ftg

redundant counterparts on the EDG 3B (or EDG 3A). Moreover, the minimum safeguards are supplemented with additional desired loads (manually loaded) such as plant investment loads as shown in Tables 3-1 and 3-2, and there is no EOG load concern. The enhanced EPS thus meets its short-term safety function for a design basis LBLOCA accident on Unit 3. Similarly, Tables 3-1 and 3-2 indicate that if a Unit 4 EOG is lost, equipment required for safe shutdown is still available.

3.3.2 LOCA on Unit 4 If the LBLOCA were on Unit 4, instead of Unit 3, the short-term kW 3-2 (for the same Load loadings similar to those given in Table Center alignments to each Unit's Train A) were calculated and result in the following:

EDG 3A kW EDG 3B kW EDG 4A kW EDG 4B kW 2344 1856 2285 1852 Thus if a LBLOCA occurs on Unit 4 instead of Unit 3, the above totals again indicate, even for the concurrent loads assumed, that each EDG kW load is still well below its continuous rating. In addition, the required short-term loads are still available even if EDG 4A (or EDG 4B) is assumed to fail. The requirement for an additional High Head Safety Injection Pump (since one is lost by failure of a Unit 4 EOG) is met by EOG 3A, with yet another HHSI Pump available if desired from EDG 3B (or 'vice-versa). The loss of the other short-term components needed is accommodated by their redundant counterparts on the EOG 4B (or 4A). Horeover, the minimum safeguards are supplemented with additional desired loads (manually loaded) such as plant investment loads as shown in the Table 3-1 and 3-2 scenarios, and there is no EDG load concern. The enhanced EPS thus meets its short-term safety function for a design basis LBLOCA accident on Unit 4 and safe shutdown of Unit 3.

3.4 ENHANCED EPS - RESULTS The above discussions are summarized as follows, to determine the worse case loading on an EOG with the enhanced EPS configuration:

. CASE 1: SIS ON UNIT 3 LBLOCA plus LOOP on Unit 3, LOOP on Unit 4; no single failures; normal (Train B) alignment; concurrent manual loads assumed:

EDG 3A kW -EOG 3B kW EDG 4A kW EOG 4B kW 1846 2159 2062 2285 3-4

"lf i(' h qt'-

k I

(4 k

CASE 2: SIS ON UNIT 4 LBLOCA plus LOOP on Unit 4, LOOP on Unit 3; no single failures; normal (Train B) alignments; same concurrent manual loads assumed:

EOG 3A kW EDG 3B kW EDG 4A kW EDG 4B kw 2016 2204 1927 2210 CASE 3: SIS ON UNIT 3 LOADS ALIGNED TO TRAIN A LBLOCA plus LOOP on Unit 3, LOOP on Unit 4; Load Center 3H and Load Center 4H aligned to Train A; concurrent manual loads assumed:

EOG 3A kw EDG 3B kW EOG 4A kW EOG 4B kW 2204 1801 2425 1922 I

CASE 4: SIS ON UNIT 4 LOADS ALIGNED TO TRAIN A LBLOCA plus LOOP on Unit 4, LOOP on Unit 3; Load Center 3H and Load Center 4H aligned to Train A; same concurrent manual loads assumed:

EOG 3A kW EDG 3B kW EDG 4A kW EOG 4B kW 2344 1856 2285 1852 Therefore, EDG 3A is most heavily loaded for Case 4 (LOOP plus Train A alignment of swing loads); EDG 3B is most heavily loaded for Case 2 (LOOP with normal Train B alignment); EOG 4A is most heavily loaded for Case 3 (LOOP plus Train A alignment of swing loads); and EOG 4B is most heavily loaded for Case 1 (LOOP with normal alignment). In other words, since the normal alignment LOOP loads are higher than the LOOP plus LOCA loads, the new EPS configuration more evenly distributes the loads on each EOG such that the normal Train B alignment of the loads for LOOP are around 2200 kW on each EDG, and if a Load Center is lost such that the loads swing onto the other EDG, the LOOP, or LOOP and LOCA, loads remain at about that level for the 3A or 4A EDG.

3.5 LONG TERN EDG LOADING 3.5.1 LOCA on Unit 3, Both Units Aligned to Train B Tables 3-1 and 3-2 indicate that for a LBLOCA on Unit 3 the required short-term equipment is available with the enhanced EPS design, with no EDG loading problems. Inspection of Tables 3-1 and 3-2 indicates that, even if one EDG on the accident Unit is assumed to fail, the required short-term equipment is still available. At least one HHSI pump is available from the 4A or 4B EDG. No equipment needs to be secured from an EOG loading standpoint.

Even though the DBA licensing basis for PTP is attainment of safe (hot) shutdown, the following discussions analyze the EOG loadings for achieving long-term cold shutdown, post-DBA. The equipment 3-5

'Lg E'

II%

fi4 r

~ KL, t

required long-term can be loaded on the accident Unit EDGs even with all short-term-required safeguards equipment still running (if required), and still maintain the EDG loads to below the EDG continuous rating. This is demonstrated on Table 3-3 and in the following discussions. Normal Train B alignment is discussed first for a LOCA on Unit 3, then a LOCA On Unit 4. The Train A alignment is briefly reviewed for resultant EDG loadings for a LOCA on Unit 3, then a LOCA on Unit 4. As per the previous Subsections, a tabulation is given for both Train alignments for the LOCA on Unit 3.

The requirement for a stable sump pH requires the following equipment to be operated within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> after the onset of the accident:

- 1 out of 3 Charging Pumps

- 1 out of 4 (2 per Unit) Boric Acid Transfer Pumps It is assumed that when boric acid injection is needed, the following equipment is also required to assure boric acid solubility:

- 1 out of 3 (Shared) Boric Acid Tank Heaters

- 1 out of 2 (Shared) Circuits for Boric Acid Heat Tracing The Boric Acid Heat Trace transformers are auto-connected, in the short-term. The other above loads on the accident Unit (Unit 3) for the long term, without securing any equipment already assumed running in the .loadings obtained previously, are depicted in Table 3-3. The loads required for shutdown and cooldown of the non-accident Unit (Unit 4) have already been assumed loaded onto the Unit 4 EDGs, and are also depicted in Table 3-3; a Spent Fuel Pit Pump is also loaded onto each Unit. The loads are as follows:

EDG 3A kw EDG 3B kW EDG 4A kW EDG 4B kW 2091 2299 2152 2285 These loads are obtained by assuming (refer to Figure 1) that a Charging Pump (125 kW each) is powered from Load Center 3A (on EDG 3A) and Load Center 3B (on EDG 3B); the Boric Acid Tank Heater "A" (15 kW) is powered from MCC 3C (on EDG 3A); the Boric Acid Transfer Pump "3A" (15 kW) is powered from HCC 3C; and the Boric Acid Transfer Pump "3B" (15 kW) is powered from HCC 3B (on EDG 3B). A Boric Acid Heat Trace circuit transformer (25 kW) is already assumed powered from HCC 3D (on EDG 3B). In addition, it is assumed that a Spent Fuel Pit Pump (90 kW) is loaded onto each Unit's "A" EDG

'(EDG 3A/4A) via LC 3C/4C. Thus an additional 245 kW is added to the pre-existing EDG 3A kW loadings (Table 3-1), and an additional 140 kW is added to the pre-existing EDG 3B kW loadings without securing any already-running equipment, and without exceeding the continuous rating of the accident Unit's EDGs (2500 kW). An additional 90 kW

. is added to the non-accident Unit's Train A diesel (EDG 4A).

3-6

I'd(

Ai t+

~l

)'/~

"tM

Tables 3-1 and 3-3 demonstrate that the most-heavily loaded EDG on the accident Unit remains below its continuous rating (2500 kW), and the enhanced EPS fulfills its safety functions in both the short and long term, even assuming the concurrent loading of engineered safety features and manual loads such as plant investment loads.

For the long-term loads needed on the accident Unit, if an EDG fails on that Unit, the loads which would have been provided by the now-failed EDG are provided by the redundant components available from the swing LC on the accident Unit, or by the redundant components on the non-accident Unit's EDGs.

For example, if EDG 3A is postulated to fail, the Boric Acid Transfer Pump "3B" is still powered from EDG 3B, as is a Charging Pump. The long-term load of the Boric Acid Tank (BAT) Heater (a shared component) is provided by using one or both of the BAT heaters normally aligned to EDG 4B, and the Boric Acid Heat Trace circuit 'transformer and the other ESF loads are powered by HCC 3D, which is still aligned to EDG 3B. If EDG 3B is postulated to fail, a Charging Pump, Boric Acid Tank Heater, and Boric Acid Transfer Pump "3A" remain powered from EDG 3A, while the Boric Acid Heat Trace circuit transformer (and ESF loads) on HCC 3D will swing to EDG 3A when the normal alignment Load Center 3D loses power and Load Center 3H automatically aligns itself to Load Center 3C.

Therefore, a single failure of EDG 3A or 38 would not affect the long-term loads needed for Unit 3, as shown on Table 3-3. The enhanced EPS thus meets its long-term safety, function for a design basis LBLOCA accident on Unit 3. Equivalent conclusions are obtained for the Train A alignment (see Subsection 3.5.3 and Table 3-4), since the accident Unit EDGs are less heavily loaded, and since the same equipment discussed above can be shown to be available.

3.5.2 LOCA on Unit 4, Both Units Aligned to Train B As noted in Subsection 3.2.2, the resultant short-term EDG kW loads for a LBLOCA on Unit 4 (plus LOOP, no single failures, normal B Train alignment) are the following:

EDG 3A kW EDG 3B kW EDG 4A kW 'DG 48 kw 2016 2204 1927 2210 The loads required long-term for the non-accident Unit (Unit 3) have already been assumed loaded onto the Unit 3 EDGs, and are repeated below, with the addition of a SFP Pump on EDG 3A.

The resulting EDG loads on the accident Unit (Unit 4) in the long-term, without securing any equipment already assumed running in the loadings 'obtained above, have been evaluated as per the discussions in Subsection 3.5.1 above and are as follows:

3-7

J~

)I

EOG 3A kW EDG 38 kW EDG 4A kW EDG 48 kw 2106 2204 2157 2380 These loads are obtained by assuming (refer to Figure 1) that a Charging Pump (125 kW each) is powered from Load Center 4A (on EDG "C"

4A) and Load Center 48 (on EDG 48); the Boric Acid Tank Heater (15 kW) is powered from HCC 48 (on EDG 48) and the Boric Acid Tank Heater "8" is powered from HCC 48 (on EOG 48); the Boric Acid Transfer Pump "4A" (15 kW) is powered from HCC 4C (on EDG 4A) and the Boric Acid Transfer Pump "48" is powered from HCC 4D (on EOG 48). A Boric Acid Heat Trace circuit transformer (25 kW) is already powered from HCC 4D (on EDG 48). In addition, it is assumed that a Spent Fuel Pit Pump (90 kW) is loaded onto each Unit's "A" EDG (EDG 3A/4A) via LC 3C/4C. Thus an additional 230 kW is added to the pre-existing EOG 4A kW loadings, and an additional 170 kW is added to the pre-existing EOG 48 kW loadings without securing any already-running equipment, and without exceeding the continuous rating of the accident Unit's EOGs (2874 kW).

For the long-term loads needed on the accident Unit (Unit 4), if an EDG fails on that Unit, the loads which would have been provided by the now-.failed EDG are provided by the redundant components available from the swing LC on the accident Unit, or by the redundant components on the non-accident Unit's EDGs. In this case, the reassignment of loads for the enhanced EPS has provided most of the long-term loads from the swing LC, as discussed below.

For example, if "48" EDG is 4A is postulated to fail, the Boric Acid still powered from EDG 48, as is a Charging Transfer Pump Pump, both Boric Acid Tank Heaters, and the Boric Acid Heat Trace circuit transformer. If EOG 48 is"4A"postulated to fail, a Charging Pump and Boric Acid Transfer Pump remain powered from EDG 4A, and the Boric Acid Tank Heater "8", and the Boric Acid Heat tracing circuit transformer (and ESF loads) will all swing to EDG 4A when their normal Load Center 4D loses power and Load Center 4H automatically aligns itself to Load Center 4C.

Therefore, a single failure of EDG 4A or 48 would not affect the long-term loads needed for Unit 4. The enhanced EPS thus meets its long-term safety function for a design basis LBLOCA accident on Unit

4. Similar to the discussions above, the enhanced EPS also meets its long-term safety functions for safe shutdown on the non-accident Unit (Unit 3).

3.5.3 LOCA on Unit 3, Both Units Aligned to Train A.

For illustrative purposes, the Train A alignment of loads on both units, and the resultant long-term EDG kW loadings, are shown in Table 3-4 for a LBLOCA on Unit 3. By the same logic used in the Train 8 alignment discussions, there is sufficient EDG capacity to mitigate the accident in the long term on Unit 3, and to safely cool down and shut down the non-acci'dent Unit 4, without any EOG loading concerns. Single failure is accommodated with the enhanced EPS design.

3-8

n, d

II '

3.5.4 LOCA on Unit 4, Both Units Aligned to Train A If the LBLOCA were on Unit 4, and the swing LCs on both Units were aligned to the "A" Train then the long-term kW loadings similar to those given in Table 3-4 (for the same Load Center 3H/4H Train A alignments) would be:

EDG 3A kW EDG 3B kW EDG 4A kW EDG 4B kW 2434 1856 2530 2007 By the same logic previously detailed, sufficient equipment is available to mitigate the accident and safely shut down the non-accident Unit, without EDG load concerns. Single failure is accommodated with the enhanced EPS design.

3.6 CONCLUSION

S Tables 3-1 through 3-4, and the kW loads shown above, demonstrate that even with the engineered safety features, desired manual loads, and the required long-term loads on both Units, plus the failure of a Load Center 3D or 4D (or alternate alignment of LC 3H/4H), there is no EDG loading problem associated with the enhanced EPS design, and the enhanced EPS can fulfill its safety function for a design basis LBLOCA on either Unit. The required equipment is available assuming a single failure of any one EDG.

After implementation of the enhanced EPS project, the emergency diesel generator (EDG) r atings (gross) will be as follows:

Unit 3 EDG's Unit 4 EDG's Continuous Rating 2500 kW

• 2874 kW (Nominal)

Short Term (2/24 Hr) 2750 kW

• 3162 kW Rating The above ratings are the kW values prior to subtraction of the EDG-run auxiliaries; i.e., before subtracting the related EDG's vent fan, air compressor, and other auxiliaries.

The enhanced EPS provides mitigation of a Design Basis Accident on one Unit and safe (hot) shutdown of the other Unit, with concurrent LOOP on both Units, for either the normal (Train B) alignment of the swing LCs or for the Train A alignment. Failure of an EDG during this DBA scenario does not affect this safety function and none of the EDGs exceeds its continuous rating.

• Per NRC's Emergency Diesel Generator Load Safety Evaluation for Turkey Point Nuclear Units 3 & 4; dated December 15, 1986, the existing EDGs can be loaded to 2750 kW with a short term continuous load limit of 2850 kW. Based upon the above manufacturer's ratings, these load limits are consistent with Regulatory Guideline 1.9, Position C.2.

3-9

1,)

f( tl 5~

pJ'~t

'c ~P

~+4 "r

4 l,~

0 TABLE 3-1 SHORT-TERN EDG kM LOADS FOR UNIT 3 LOOP PLUS LOCA TMO-UNIT OPERATION MITH SMING LOADS ALIGNED TO B TRAIN (NOTE: C(NPONENT kM LOADS INCREASED FOR CONSERVATIQI)

NO SINGLE FAILURE) REFER TO SECTION 3.0 DISCUSSIONS U3: LON + SIS U4: LOOP ONLY COeCWENTS ( ) EDG 4A EDG CB RENARKS HHSI 305 305 305 305 Only trro HHSI pmps required for accident Unit RHR 225 225 (225) (225) Only one RHR pap per Unit required CS 225 225 N/A N/A Only one CS pmp required on accident Unit CCM 380 380 380 380 Only one CCM purp required per Unit I CM 270 270 270 270 Only one ICM punp required per Unit Noraal Contairaent Coolers N/A N/A (160) (160) Loss of parer to LC 4D suings NCC 4B (NCC 4D)

Eaergency Contairaent Coolers 25 50 N/A N/A Loss of pow.r to LC 3D srrings ECC 3B (NCC 30)

Eaergency Contairaent Fi L ters 65 130 N/A NIA Loss of parer to LC 3D suings ECF 3B (NCC 3D)

Battery Chargers 25 75 25 75 NCCs 38, 3C, 30, 48, CC, 40 parer 8 battery chargers (1 per DC bus req'd)

Charging Prap N/A N/A (125) (125) Neatly loaded for hot shutdarn RCS inventory control.

Pressurizer Heaters N/A N/A (150) (150) Nanually Loaded for hot shutdarn RCS pressure control.

Turbine Loads see "plant investaent" see "plant investaent" Nanual ly loaded.

Eaergency Lighting XFNR 0 20 0 20 loss of LC 3D or 4D srrings Earg Ltg XFNR 312 (NCC 3D) or 412 (NCC 40)

Control Ram AC 0 60 0 30 loss of LC 3D or 40 srrings CRAG C (NCC 3D) or CRAC B (NCC 40)

BA Heat Tracing XFNR 0 25 0 25 Loss of LC 3D or 40 suings XNR 3X313 (NX 3D) or 4X416 (NCC 40)

EDG Auxiliaries 30 30 115 115 NXs 3A (EDG 3A) ~ 3K (EDG 38), CJ (EDG 4A), CK (EDG 4B)

Niscellaneous Loads 35 35 35 35 Est lasted Load Center Transforaer Losses 30 30 30 30 Varies depending on load; full load value used Battery Rooa AC E16D 8 E16F (0) (25) (0) (25) Loss of LC 3D or 4D srrings Batt Ra AC E160 (NCC 3D) or E16F (NCC4D)

H2 Analyzer Related 10 10 10 10 One Train on accident Unit required Carputer Ra/Cabte Sprdg Roas AC (0) (50) ( 0) (50) Nanuatly loaded; one req'd. Loss of LC 3C or 4C suings AC 8 or AC A Boric Acid Transfer Pmp N/A N/A (15) (15) Nanually loaded; one req'd.

Boric Acid Tank Htr NIA N/A (0) (30) Not required short-tera; loss of LC 4D srrings BAT Htr B (NCC 4D)

Aux Bldg/EER HVAC 0 68 0 68 Process Load; Loss of LC 3D or 40 srrings HVAC (NCC 3D/4D) 4.16 kV Srrgr/EDG Ra HV 46 46 42 42 Process Load for Srrgr and EDG roans Plant Investaent Loads (est.) 175 ~100 ~175 ~100 Plant investaent Loads incl turbine loads, both aarvjat 8 process-auto.

TOTAL KM LOADING: 2159 NOTES: 1) Security Building Transforser toad not sham due to Security Upgrade Project.

EDG Contirarous Rating (naa.) 2500 2874 2) Parentheses denote sirat load kM values.

t,i I'

g P

pl Vy I

n,

(NOTE:

NO TA 2 SHORT-TERN EDG kM LOADS FOR UNIT 3 LOOP PLUS LOCA TMO-UNIT OPERATION MITH SMIHG LOADS ALIGNED TO A TRAIN COHPOHEHT kM LOADS INCREASED FOR COHSERVATISN)

SINGLE FAILURE; REFER TO SECTION 3.0 DISCUSSIOHS I

U3r LOOP + SIS U4: LOOP DHLT C(HCPOHENTS ( ) ( ) EDG 3A EDG 3B EDG 4A EDG 4B REHARKS HHSI 305 305 305 Only tMo HHSI pcmps required for accident Unit RHR 225 (225) (225) Only one RHR pcmp per Unit required 225 N/A N/A Only one CS pcmp required on accident Unit 380 380 380 Only cne CCM pcrrp required per Unit 270 270 270 Only one ICM pcmp required per Unit Normal Contairmnt Coolers H/A N/A (240) (80) Loss of LC 4C sMings HCC 4B (NCC 40)

Emergency Contairment Coolers 50 25 N/A N/A Loss of LC 3C sgings ECC 38 (HCC 3D)

Emergency Contaiaaent Filters 130 65 N/A N/A Loss of LC 3C siings ECF 3B (NCC 30)

Battery Chargers 75 25 75 25 NCCs 38, 3C, 30, 48, 4C, 40 poMer 8 battery chargers (1 per DC bus req'd)

Charging Pcmp H/A N/A (125) (125) Nanually loaded for hot shutdorn RCS inventory control.

Pressurizer Heaters N/A N/A (150) (150) Nanually loaded for hot shutdan RCS pressure control.

Turbine Loads see "plant investment" see "plant Investment>> Hanuaily loaded.

Emergency Lighting XFHR 20 0 20 0 Loss of LC 3C or 4C suings Emrg Ltg XFHR 312 (NCC 3D) or 412 (HCC 40)

Control Roaa AC 30 30 30 0 Loss of LC 3C or 4C sMIngs CRAC C (HCC 3D) or CRAC 8 (NCC 40)

BA Heat Tracing XFHR 25 0 25 0 Loss of LC 3C or 4C srrings XFNR 3X313 (HCC 30) or 4X416 (NCC 40)

EDG Auxiliaries 30 30 115 115 NCCs 3A (EDG 3A) ~ 3K (EDG 38) ~ 4J (EDG 4A), 4K (EDG 4B)

Niscellaneous Loads 35 35 35 35 Estimated Load Center Trrnsformer Losses 30 30 30 30 Varies depending on load; full load value used Battery Ram AC E160 8 E16F (25) (0) (25) (0) Loss of LC 3C or 4C sMings Batt Rm AC E16D (HCC 3D) or E16F (NCC 4D)

H2 Analyzer Related 10 10 10 10 One Train on accident Unit required Cosputer Rm/Cable Sprdg Roas AC (50) (0) (50) (0) Nanually loaded; one req'd. Loss of LC 3C or 4C srrings AC A or AC B.

Boric Acid Transfer Pcmp N/A N/A (15) (15) Nanually loaded; one req'd.

Boric Acid Tank Htr N/A N/A (15) (15) Hot required short-tera; loss of LC 4C sMings BAT Htr 8 (HCC 4D)

Aux Bldg/EER HVAC 68 0 0 Process Load; Loss of LC 30 or 4D srrings HVAC (NCC 30/4D) 4.16 kV SMgr/EDG Rm HV 46 46 42 42 Process Load for SMgr and EDG rooms Plant Investment Loads (est.) ~100 ~175 ~100 Plant invstant loads incl turbine loads, both manual 8 process-auto.

TOTAL KM LOADING: 1801 2425 1922 NOTES: 1) Security Building Transformers load not shan due to Security Upgrade Project.

EDG Continuous Rating (nom.) 2874 2874 2) Parentheses denote marwai load kg values.

E.

k, 75

'i>

l

TAB -3 LONG-TERM EDG klr LOADS FOR UNIT 3 LOOP PLUS LOCA TWO-UNIT OPERATION lllTH SUING LOADS ALIGNED TO B TRAIN (NOTE: COMPONENT kM LOADS INCREASED FOR CONSERVATISM)

NO SINGLE FAILURE; REFER TO SECTIOH 3.0 DISCUSSIOHS lG= L(XX + SIS U4 LOOP ONLT carpoNENTs (1) EDG 3A EDG 3B EDG 4A EDG 4B REMARKS HHSI 305 305 305 305 Only tMo HHSI purps required for accident Unit RHR 225 225 (225) (225) Only one RHR pap per Unit required 225 225 N/A N/A Only one CS purp required on accident Unit 380 380 380 380 Only one CCM pap required per Unit 270 270 270 270 Only one ICM pap required per Unit Normal Contairmnt Coolers K/A K/A (160) (160) Loss of LC 40 srrings NCC 4B (NCC 40)

Emergency Contairarent Coolers 25 50 N/A N/A Loss'of LC 3D sMings ECC 3B (NX 3D)

Emergency Contairarent Fitters 65 130 N/A K/A Loss of LC 3D srrings ECF 38 (NX 3D)

Bat tery Chargers 25 75 25 75 NXs 38, 3C, 30, 48, 4C, 4D parer 8 battery chargers (1 per DC bus req'd)

Charging Prap (125) (125) (125) (125) Manually loaded for HSD and boration; one per lsrit required Pressurizer Heaters N/A N/A (150) (150) Nanually Loaded for hot shutdarn RCS pressure control.

Turbine Load see "plant investment~ see "plant investment" Nanual ly Loaded.

Emergency Lighting XFNR 0 20 0 20 Loss of LC 3D or 4D swngs Emrg Ltg XFNR 312 (NCC 3D) or 412 (NCC CD)

Control Roaa AC 0 60 0 30 Loss of LC 30 or 40 srrings CRAG C (NCC 3D) or CRAC B (NCC 40)

BA Heat Tracing XFNR 0 25 0 25 Loss of LC 30 or 40.srrings BA Ht Tr XFNR 3X313 (NCC 30) or 4X416 (NCC 4D)

EDG Auxiliaries 30 30 115 115 NCCs 3A (EDG 3A) ~ 3K (EDG 38) ~ 4J (EDG 4A) ~ 4K (EDG 4B)

Ni seel l aneous Loads 35 35 35 35 Estimated Load Center Transforaer Losses 30 30 30 30 Varies depending on load; full load value used Battery Roas AC E160 t E16F (0) (25) (0) (25) Loss of LC 3D or 4D sMings Batt Rm AC E160 (NCC 30) or E16F (NCC 40)

H2 Analyzer Related 10 10 10 10 One Train on accident Unit required Spent Fuel Pit Pp (90) (90) Spent fuel cooling aay be needed rrithin about 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> Ccaputer Rrrr/Cable Sprdg Rum AC (0) (50) (0) (50) Nanually loaded; one req'd. Loss of LC 3C or 4C srrings or AC A or AC B Boric Acid Transfer Pump (15) (15) (15) (15) Nanually loaded; one req'd.

Boric Acid Tank Htr (15) (0) ( 30) Kot required short-tera; Loss of LC 4D srrings BAT Htr B (NCC 40)

Aux Bldg/EER HVAC 0 0 68 Process Load; l.oss of LC 30 or 4D sMings HVAC (NCC 30/40) 4.16 kV Sugr/EDG Ra HV 42 42 Process Load for SMgr and EDG roams Plant Investment Loads (est.) ~175 ~175 ~lm Plant invstmt Loads incl turbine loads, both aarxrat 8 process-auto.

TOTAL KN LOADING: 2152 NOTES: 1) Parentheses denote marxrat load kll values.

(including aarxrat loads) 2) Loads conservatively shorrn for attairmcnt of cold shutdorrn.

EDG Cont irwous Rating (nan.) 2874 2874 4

f 4

fo oi 21

TABLE LONG-TERN EDG kM LOADS FOR UHIT 3 LOOP PLUS LOCA TMO-UNIT OPERATIOH MITH SMING LOADS ALIGNED TO A TRAIN (NOTE: COHPONENT kM LOADS INCREASED FOR CONSERVATISH)

NO SINGLE FAILURE; REFER TO SECTION 3.0 DISCUSSIOHS LG: LCXH5 + SIS U4: LCXH5 ONLY CCIPONENTS EDG 4A EDG 4B RBCARKS HHSI 305 305 305 305 Only tao HHSI Ixaps required for accident Unit 7 225 (225) (225) Only cue RHR Ixap per Unit required 225 225 N/A K/A Only one CS pcap required on accident Unit

'80 380 380 380 Only one CCM peep required per Unit 270 270 270 270 Mty one ICM peep required per Unit Noraat Contairmnt Coolers N/A N/A (240) (80) Loss of LC 4C slcings KCC 48 (NX 4D)

Eaergency Con'taIMcnt Coolers 50 25 N/A N/A Lees of LC 3C sMings ECC 38 (HCC 3D) '-

Eaergency Contaicvaent Fi Lters 130 65 N/A N/A Loss of LC 3C suings ECF 38 (NCC 3D)

Battery Chargers 75 25 75 25 NXs 38, 3C, 3D, 4B, 4C, 40 pouer 8 battery chargers (1 per DC bus req'd)

Charging Pcap (125) (125) C125) (125) Nanually loaded for HSD and boration; one per mit required Pressurizer Heaters N/A N/A (150) (150) Nanually loaded for hot shutdoun RCS pressure controt Turbine Loads see Mptant investaent" see Mptant investaent55 Hanual ty loaded Eaergency Lighting XFHR 20 0 0 Loss of LC 3C or 4C scavIs Earg Ltg XFIR 312 (NX.30) or 412 (NX 4D)

Control Rocxa AC 30 30 30 0 Loss of LC 3C or 4C suings CRAG C (HCC 30) or CRAC 8 (HCC 40)

BA Heat Tracing XFNR 0 25 0 Loss of LC 3C or 4C suings BA Ht Tr XFHR 3X313 (IHX 3D) or 4X416 (NX 4D)

EDG Auxiliaries 30 30 115 115 IHXs 3A (EDG 3A), 3K (EDG 38) ~ 4J (EDG 4A) ~ 4K (EDG 48)

Niacct leneous Loads 35 35 35 35 Estiaated Load Center Trensforaer Losses 30 30 30 30 Varies depending on Load; fuLL load value used Battery Rcxm AC E160 8 E16F (25) (0) (25) (0) Loss of LC 3C or 4C su3ngs Batt Ra AC E16D (NCC 3D) or E16F (NX 4D) 82 Analyzer Related 10 10 10 10 ika Tra3n on accident Unit required Spent Fuel Pit Pcap C90) (90) Spent fuel cool3ng any be needed lc3thln about 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />

~er Ra/Cable Sprdg Roca AC (50) CO) (50) (0) Nanuatty loaded; one req'd. Loss of LC 3C/4C slclngs AC A/8 Boric Acid Transfer Puap (15) (15) (15) (15) Nanually loaded; one req'd.

Boric Acid Tank Htr (15) (15) (15) Not required short-tera; loss of LC 4C suings BAT Htr 8 (NX 40)

Aux Bldg/EER HVAC 0 68 0 Process Load; Loss of LC 30 or 4D su3ngs HVAC (NX 3D/4D) 4.16 kV SMgr/EDG Ra HV 42 42 Process Load for SMgr and EDG roaas Plant Investaent Loads (est.) ~115 ~7M ~175 ~1M Ptent invstmt loads incl turbine loads, both aacxcat a process-auto.

TOTAL KM LOADING: 2449 1941 2515 1922 NOTES: 1) Parentheses denote aanual load kM values.

2) Loads conservatively shock for attairment of cold shutdown.

EDG Continuous Rating (nan.) 2874 2874

W4

(

4.0 SE UENCER ENHANCEMENTS Voltage drop calculations performed during the design of the swing.

Load Centers 3H/4H and the repowering of equipment associated with the associated Motor Control Centers 30/40, indicated that the load sequencing could be enhanced by modifying the sequence of loads and adding additional load blocks during the one-minute duration when the auto-connect loads are placed on the EDGs. In addition, it was advantageous from voltage drop considerations to

~T dgf '~ \ have the loads also sequence onto the buses for the scenario of an SI with offsite power available. Finally, certain single failure scenarios require the redundant swing loads to be loaded onto an EDG following the normal sequencing of the other ESF and hot shutdown loads. A description of these changes is provided below.

4.1 'XISTING LOAD BLOCKS

'I For a LOOP/LOCA on Unit 3, the Updated FSAR Table 8.2-3 (Revision 6, 7/88) is the basis for the following worst case Engineered Safeguards Features loading on an EDG with four load blocks in the existing sequencer design (see next page). In addition, for a LOOP on Unit 4, the required Unit 4 loads will sequence on after the Unit 3 ESF loads.

4-1

pr I

1 lh h

I'

SECONDS AFTER ENGINE START BLOCK SIGNAL MAJOR E UIPMENT ~LOAD kW 16 LC Transformer Losses 9 Lighting 17 Motor Operated Valves

• 30 Miscellaneous Loads 17 Heat Tracing 20 Control Room Air Conditioners 54 Battery Chargers 69 Battery Room Air Conditioner 22 Step Load 238 18 High Head SI Pumps 604 Residual Heat Removal Pump 201 Step Load 805 26 Component Cooling Water Pump 365 Emergency Containment Cooler 23 Emergency Containment Filter 55 Containment Spray Pump 219 Step Load 662 33 Intake Cooling Water Pump 265 Emergency Containment Cooler' 23 Emergency Containment Filter(a'tep 55 Load 343 (Sequencer Times Out)

TOTAL'W 2048

• Approximate total kW load for multiple short-term, small loads This represents the worst case loading scenarios for one EDG Assumes accident on Unit 3; loads only apply for this case. If an SIS on Unit 4, only one cooler and filter starts per EDG.

4-2

4 kg

4.2 ENHANCED LOAD BLOCKS For the enhanced design, the following ESF load blocks were designed, for an SI with LOOP scenario. (The kW loads shown here are higher than design for conservatism.) The loads are shown for EDG 3B, assuming the normal alignment of LC 3H/MCC 3D to the B Train:

SECONDS AFTER ENGINE START L ~LOAD kW

"'5.5 LC Transformer Losses 30 Lighting Transformer 20 Hotor Operated Valves* 30 Niscellaneous Loads Battery Chargers '535 Heat Tracing 20 Step Load 210 18 High Head SI Pump 305 Residual Heat Removal Pump 225 Step Load 530 26 Emergency Containment Cooler 25 Containment Spray Pump" 225 Step Load 250 33 Intake Cooling Water Pump ** 270 Emergency Containment Cooler 25 Step Load 295 40 Component Cooling Water Pump Step Load 380

.54 59 Emergency Containment Emergency Containment (Sequencer Times Out)

Filter Filter '5 Step Load Step Load 65 65 65 TOTAL kW 1795

• Approximate total kW load for multiple short-term, small loads

• • If Containment Spray Pump is not required to start by the 3rd load block, upon the receipt of a High-High Containment Pressure Signal it will start after sequencing is complete.

• • • single failure (discussed below) delays auto-transfer of If LC/HCC, the ECCF will load onto the EDG at 72 seconds after the engine start signal.

NOTE: For LOOP only scenario,. the above sequence times apply for non-ESF loads 4-3

Ffg 0

yP yl C

The times shown in 4. I and 4.2 are total elapsed times from receipt of a LOOP signal. Note that an additional one second is required for the undervoltage relay to initiate sequencing subsequent to detection of the LOOP.

The assignment of a High Head Safety Injection Pump to each EDG in the enhanced design deletes about 305 kW from the second load block in 'each EDG loa'ding, and the assignment of seven load blocks instead of four load blocks more evenly distributes the loading, and thus allows the voltage to recover more readily.

4.3 SEQUENCING LOADS ONTO BUSES FOR SI WITH OFFSITE POWER AVAILABLE If an SI occurs with offsite power available, the enhanced EPS design utilizes the sequencers to sequence the loads onto the buses, with the same load blocks shown in Subsection 4.2 above, but without the 15-second delay for the EDGs to come up to speed and voltage.

That is, the load block times shown in 4.2 occur 15 seconds earlier than for the SI with LOOP scenario. Note that any equipment operating before the receipt of the SI signal will remain operating.

Since the Engineered Safety Features are loaded onto the buses earlier than for the limiting case of LOOP plus SI, the FSAR accident analyses (DBA with LOOP) remain the bounding analyses for accident consequences because of the delay to initiate the ESF equipment.

ENHANCED SEOUENCER DESIGN TO NEET SINGLE FAILURE CRITERIA The new sequencers logic and loading blocks were designed to meet single failure criteria. The design of the load blocks for the enhanced EPS consider accident response requirements to ensure plant safety, engine loading performance and single failure. The repowering of loads from the existing HCC 0 (relabeled MCC 30) and timing of the load blocks have been considered. The design includes several features which ensure that the consequences of component failures do not impact plant safety:

I. The swing load centers/motor control centers are isolated and not powered if the charging pump breaker fails to strip upon command.

2. Loading of the swing load center is delayed until after the seventh load block to accommodate any failure, during sequencing, which causes loss of the train to which the LC was initially aligned.

The loading sequence of the enhanced EPS is designed to,ensure it does not adversely affect the existing FSAR accident analyses. The addition of new load blocks and the delayed loading (compared to the existing loading design) of certain ESF equipment have been considered in the design. At a minimum, all of the equipment assumed in these analyses is available and loaded on the EDGs. With the.EDG load margin available with the enhanced EPS, the capability exists to power additional loads. An evaluation is being finalized which will verify that the proposed changes to the EDG loading 4-4

~rA scheme do not adversely impact the current design basis accident analyses for Turkey Point Units 3 and 4. A preliminary review (Reference 20), of the proposed changes, by Westinghouse indicates that there will be no adverse impact on the FSAR analyses of record.

4-5

h I

l' 4

5.0 FAILURE MODE AND EFFECTS ANALYSES FHEA In addition to postulating various single failures during the accident scenarios described in the previous section, the design of the enhanced EPS system was subjected to a rigorous analysis of failure modes during system operation, to demonstrate that the system safety function can still be accomplished even with a single failure. The enhanced EPS was subjected to a comprehensive FHEA, the key elements of which are summarized below, from which it is concluded that the enhanced EPS can perform its safety functions for all modes of plant operation.

The detailed FHEA is approximately 500 pages long, and is not included in this Safety Analysis for the sake of brevity. The FHEA for the electrical system components and individual loads was used to identify potential failure modes from proposed relocation (repowering) of the shared, or common, loads which presently are located on HCCs 3A, 4A or D. Where the FMEA identified such problems, the loads were assigned to the proper HCC such that a DBA and single failure does not result in the loss of the minimum required equipment. Figure 1 depicts the design which evolved from the FHEA studies and shows selected repowered loads (indicated by an asterisk on Figure 1).

The FHEA also considered the scenario of one Unit at power and the other Unit in MODE 5 or 6 with one train out-of-service (OOS). An accident (LOCA) plus a single failure was assumed on the Unit at power. The FHEA was used to confirm that the proposed EPS Technical Specifications (NRC proof and review, dated 5/4/90) are appropriate for potential EPS configurations during plant operation, includi'ng equipment OOS conditions (e.g., for a Unit in MODES 5 or 6 with only one EDG required to be OPERABLE). The EPS Technical Specifications are being submitted to NRC, as a separate package, for NRC review.

5.1 FMEA OF THE ONSITE AC EMERGENCY POWER SYSTEM Operation of the enhanced EPS and its components, including initiating signals, sequencers, breakers, etc. is degcribed in the June 1988 submittal. A review of various failure modes and their effect on the EPS safety functions was performed assuming LOOP, and the accidents discussed in Section 3.0 above (i.e., assuming that both Units are at power, an accident occurs, and dual-Unit LOOP is concurrent with the accident). This scenario is the worst-case condition from an EDG loading standpoint. The FHEA shows that no single failure of the onsite EPS (i.e., sequencers, EDG and AC electrical distribution system components) will preclude meeting the enhanced EPS design bases for LOOP and LOCA, or LOOP only.

5-1

II

5.2 FHEA OF THE 125V DC ENHANCEMENTS The operation of the enhanced 125V DC components, including initiating signals, breaker controls, etc. is described in the June 1988 submittal. A review of various failure modes and their effect on the enhanced 125V DC safety function was performed assuming LOOP, and then the accidents discussed above (i.e., assuming that both Units are at power, an accident occurs, and dual-Unit LOOP is concurrent with the accident). This scenario is the worst-case condition from an EDG loading standpoint. The FHEA shows that no single failure of the 125V DC enhancements will preclude meeting the enhanced EPS design bases for LOOP and LOCA, or LOOP only.

5-2

L p

4

, l ill

6.0 PROBABILISTIC EVALUATION

6. 1 PURPOSE AND SCOPE e

The purpose of this section is to provide a quantitative probabilistic evaluation of the enhanced emergency diesel generator configuration for Turkey Point Units 3 and 4. This evaluation was performed both for the existing EDG configuration and the enhanced EDG configuration to allow a comparison of the two configurations. (The terms 4-KV and 4. 16 KV are used interchangeably in this evaluation).

Three different measures were used to evaluate and compare the two EDG configurations. These measures were

1. The probability of AC power availability on the 4. 16-KV buses given that a loss of offsite power initiating event has occurred.

2. Total frequency of the failed 4KV-Bus states that results from the loss of offsite power initiating event. In this evaluation the following items were taken into consideration:

A. Operational state of both units, e.g., power operation, shutdown, design basis accident with safety injection signal.

B. Availability states of 4-KV buses, e.g., AC power available on all, some, or none of the four emergency buses.

3. Evaluations considered AC power recovery for both designs, Unit crosstie capability with the enhanced design, and considered a conservative probability of operator error for either design.

6. 2 QUANTITATIVE EVALUATION METHODOLOGY A quantitative evaluation of a system can be done in many different ways, particularly when the system mission success criteria depends upon the operating status of the two nuclear units, as well as the nature of the initiating event challenging the system. For this evaluation, two different measures were used to evaluate and compare the two EDG configurations.

The first measure is the probability of being in a particular 4-KV bus state, where a bus state is defined in terms of the availability of AC power on given combinations of the 4. 16-KV buses following a loss of offsite power (LOOP) event.

The second measure is the total frequency of failed 4KV-bus states following a LOOP event, considering the different status of each of the two units (e.g., in power operation, shutdown, or DBA with SI).

The methodology for quantification of each of the two measures -is described in the following sections.

6-1

0 Rl gll I

t 4

IJ

6.2. 1 4-KV Bus State Probabilities The EDGs and their associated support and distribution systems're designed to respond to a loss of offsite power event which affects both nuclear power units. The EDGs supply onsite emergency AC power to the four (4) emergency buses, switchgear 3A, 3B, 4A, and 4B. The availability of AC power on these buses is a quantity which provides for an evaluation and comparison of the different EDG configurations.

After a LOOP event, the four emergency buses may be in one of sixteen 4-KV Bus states, as displayed in Table 6-1. A 4-KV Bus state is any combination of the four (4) emergency buses having and/or not having AC power after the LOOP event. for example, the first 4-KV Bus state has AC power on all four 4-KV emergency buses, whereas the 16th state has no AC power on any of the four 4-KV buses.

The conditional probabilities of these states (P,) after a LOOP event add up to 1.0. The four emergency buses at the two units have to be in one of these 16 states. Also, these states are mutually exclusive; they can not exist at the same time.

It is desirable that P be close to 1.0 and the probability of being in bus states 6, 7, and 12 through 16 be low. The 4KV-bus states 6, 7, and 12 through 15 correspond to single unit blackout, and state 16 is the station blackout at both units. The probabilities for the 16 states can be calculated by using fault tree modeling techniques, for each EDG configuration. Then the calculated probabilities can be compared for each bus state for the two configurations (e.g., compare P,. (existing configuration) with P, (enhanced configuration), etc.).

6.2. 1. 1 Procedure for 4-KV Bus State Evaluation The following procedure was utilized to evaluate EDG configurations using 4-KV Bus state probabilities as the measure of evaluation:

Fault tree models for the 4-KV Bus states, for the existing EDG configuration were constructed. Included in the models were random failures, common cause failures, operator action failures (if any),

unavailability due to test and maintenance of the EDG's, failure of the distribution and support systems, etc.

2. A data bank using generic and plant specific data to quantify failures modeled in-the fault trees was prepared.

3. Fault trees were quantified to obtain the failure probabilities of each bus state.

4. The dominant failure combinations and importances of components'were identified and listed.

5. Results from the analysis are documented.

6-2

+qC V

6. Steps I, 3 and 4 were repeated for the enhanced EDG configuration.

(The data bank of Step 2 was assumed to be applicable to both configurations.) Document the analyses and results.

7. Conclusions as to the acceptability of the bus state probabilities for each configuration were determined.

8. A comparison of the two EDG configurations from the bus state probability point of view was then performed 6.2.2 Estimation of the Total Frequency of Failed 4KV-Bus States The total frequency of failed "4KV-bus states was the second measure used to quantitatively evaluate and compare the two EDG configurations. To quantify this frequency due to the LOOP event for each EDG configuration, the following multistate model was used:

The frequency f can be formally written as a product of three terms:

LOOp) ( plant state) ( bus state)

The first term refers to the plant specific initiating event frequency for the LOOP event, expressed as the number of LOOP events affecting both units per calendar year.

The second term refers to the probability of the two units being in a given plant state. A plant state is defined as the status of a unit during the LOOP event. The following three possibilities have been chosen as limiting conditions for the plant status:

l. A unit is in power operation (PO) (with probability PPO).

2. A unit is in cold shutdown (SD) (with probability PSD).

3. A unit is in DBA with SI (SI) (with probability PSI).

The probabilities of these plant states for each unit are calculated.

Note that these plant states are assumed to make up all of the possible states that a unit can be in. Thus these probabilities sum up to 1.0 for each unit. Any other conceivable intermediate state is to be classified as one of these three states and its probability is to be added to that of the proper state.

Based on the above classification, a total of 9 two-unit plant states can be defined, as shown in Table 6-2.

The third term (P ) refers to the probabilities of failed 4-KV bus states, as descrikQ III kNctioo 6.2.1. These probabilities are calculated by using fault tree analysis results. Success and failure for a bus state are determined for each plant state.

6-3

1)

$)

The details of the calculations and the determinations are provided in Westinghouse Report "guantitative Evaluation of Enhanced Emergency Diesel Generator Configuration for Turkey Point Nuclear Power Units 3 and 4-guick-Look Assessment", 'and in the FPL Reliability and Risk Assessment Group's "PTN EPS guick-Look Assessment". The:overall results are presented below.

6.3 RESULTS When the power recovery capability of both the enhanced EPS inter-Unit crosstie and the existing system capability are modeled, the following results are obtained for the failure frequency (per year) of the 4 kV buses to provide AC power:

EXISTING ENHANCED Events With Safety Injection: 1.2 E-07 8.7 E-08 Events Without Safety Injection: 2.1 E-04 6.7 E-06 Independently of the above calculations (i.e., not taking credit for power recovery), when the operator error rate is conservatively assumed as a probability of 0. I (due to load management and'perator action requirements), the 'failure frequencies (per year) are calculated as:

EXISTING ENHANCED Events With Safety Injection: 3.2 E-07 2.3 E-07 Events Without Safety Injection: 2.1 E-03 6.7 E-05 As expected, for both cases studied above, the events with safety injection are low-probability events and therefore the failure frequency is also low. The existing and the enhanced designs both provide a low frequency for failed bus states for events with SI, with the enhanced design resulting in relatively "better" values (considering the already low frequencies calculated).

From a plant safety standpoint, the events without safety injection (i.e.,

the loss of offsite power events, with various bus failure states calculated) are a better indicator of how the enhanced design compares to the existing design,'ince events without an accompanying SI are more probable. As indicated, above, the enhanced design provides a more reliable design in-either of the two cases considered.

6-4

If the more-likely events without SI are considered, even without allowing for AC power recovery and without a conservative value attributed to operator error, the failure frequencies (per year) of the 4-KV buses to provide AC power are as follows:

EXISTING ENHANCED Events Wi thout Safety Injecti on: 2.1 E-03 6.7 E-05 Clearly, the enhanced EPS design provides a more reliable system which is better able to cope with various loss of offsite power scenarios. An enhanced capability is thus provided to mitigate relatively high-probability events.

6-5

"F+,

'E

TABLE 6-1 DEFINITION OF 4-KV BUS STATES After a LOOP event, the site may be in various 4-KV Bus states. Each bus state refers to availability of AC power on a different combination of 4-KV buses. The following bus states may exist:

AC Power On Emer enc Bus State ¹ 3A 3B 4A 4B Probability 'omments

1. Pl

2. Y P2

3. Y P3

4. Y P4

5. N P5

6. Y P6 Single Unit Blackout

7. N P7 Single Unit Blackout

8. Y P8

9. P9

10. P10 Pl 1

12. P12 Single Unit Blackout

13. P13 Single Unit Blackout

14. P14 Single Unit Blackout

15. P15 Single Unit Blackout

16. P16 Station Blackout Sum of probabi 1 i ti es 1.0 Y = Yes; there is AC power on the emergency bus.

N No; there is No AC power on the emergency bus.

TABLE 6-2 DEFINITION OF PLANT STATES Plant Unit 3 Unit 4 State ¹ Condition Condition Probabilit Comments PO PO PS1 States 1-4 refer to SI PO PS2 both units being in PO SI PS3 operation SI SI PS4

'PO SD PS5 States 5-8 refer to SD PO PS6 one unit being in SI SD PS7 operation; the other SD SI PS8 in standby SD SD PS9 State 9 refers to both units being in standby Sum o f probabi i t i es 1 - 1. 0 PO Power Operation SD Cold Shutdown SI DBA with SI

I

<~v l(

7.0 CONCLUSION

S The information presented in this Safety Analysis demonstrates that the

'enhanced EPS provides additional installed capacity at the Turkey Point Plant such that the design basis accident of LOOP, plus a LOCA on one Unit, plus the single failure of an EDG, is mitigated with 3 EDGs available. The 3 EDGs can be automatically loaded and manually loaded with the required loads for accident mitigation on one Unit and the achievement of safe hot shutdown on the non-accident Unit. In addition, the EDG loading capacity available for the design basis accident affords sufficient capacity for manual loading of the loads desired in the long-term recirculation phases of the accident or in the transition to cold .

shutdown for the non-accident Unit.

The Failure Modes and Effects Analyses, performed for the enhanced design, demonstrate that the minimum equipment required to mitigate the design basis accidents described in the FSAR is readily available with the enhanced EPS configuration, even assuming a single failure of an EDG to start. (In fact, more than the FSAR minimum equipment is often available for most accident scenarios.) Thus, the accident analyses in the FSAR remain valid as bounding analyses and the accident analyses results are not affected (i.e., still meet applicable regulatory requirements) as a result of reconfiguring the EPS by this enhancement project.

From a probabilistic standpoint, the enhanced system provides a reduction, in the overall failure frequency of the 4-KV buses to provide AC power in the case of the LOOP and LOOP with SI events, as compared to the existing configuration. Hence, overall plant safety as measured by the availability of emergency power to the plant safety buses is improved under the enhanced EPS configuration.

7-1

I,

~

p I,

gf tg

8.0 REFERENCES

Updated FSAR, Revision 7, 7/89

2. FPL Letter to NRC, L-88-269, dated June 23, 1988, "Emergency Power System (EPS) Enhancement Project," submitting EPSER

3. FPL Letter to NRC, L-88-454, dated October 19, 1988, outlining New EDG Testing and requesting Waiver of 300-Start and Load Tests FPL Letter to NRC, L-88-510, dated December 14, 1988, outlining the 10CFR50.59 Licensing Approach

5. NRC Letter to FPL, dated January 6, 1989, "Turkey Point Units 3 L 4 -Request for Additional Information on Emergency Power System Enhancement Project"

6. FPL Letter to NRC, L-89-54, dated February 24, 1989, submittal of first set of FPL Responses to NRC's 1/16/89 RAIs

7. FPL Letter to NRC, L-89-107, dated March 20, 1989, submittal of remaining FPL Responses to NRC's 1/16/90 RAIs

8. FPL Letter to NRC, L-89-124, dated April 3, 1989, submitting EPSER Supplement No 1: Testing

9. FPL Letter to NRC, L-89-144, dated April 17, 1989, submitting 10CFR50.63 Information on PTP Station Blackout 10 NRC Letter to FPL, dated March 16, 1990, "Turkey Point Units 3 L 4-Request for Additional Information on Emergency Power System Enhancement Project" FPL Letter to NRC, L-90-140, dated April 16, 1990, first submittal

'2.

of FPL Responses to NRC's 3/16/90 RAIs FPL Letter to NRC, L-90-160, dated May 2, 1990, second submittal of FPL Responses to NRC's 3/16/90 RAIs 13.. Westinghouse Letter to FPL, FPL-86-561 dated March 3, 1986, "Emergency Operating Procedures/Emergency Diesel Generator Loading"

14. JPN-PTN-SENJ-89-140, Revision 0, "PTN EPS guick-Look Assessment",

January 1990

15. Westinghouse Letter to FPL, NS-RMOI-PRRA-244 dated December 27, 1988 8-1

V,I k'

16. NUREG/CR-2728, "Interim Reliability Evaluation Program Procedures Guide" (IREP), January 1983.

17. NUREG/CR-4550, Analysis of Core Damage Frequency from Internal Events: Methodology Guidelines, Volume 1, September 1987.

18. Westinghouse "quantitative Evaluation of Enhanced Emergency Diesel Generator Configuration for Turkey Point Nuclear Power Units 3 8 4-guick Look Assessment", December 1988.

19. FPL Letter to NRC, L-86-256 dated June 16, 1988, "Reportable Event 85-42 (Revision 1), Turkey Point Unit 3, Date of Event: December 14, 1985 (original date), Emergency Diesel Generator Loading".'0.

Westinghouse Letter to FPL, FPL-90-611 dated May 7, 1990, "Assessment for Proposed Diesel Loading Schemes"

21. Ebasco letter No. PTP-90-139 Transmittal of 4kVAC, 480VAC, and 125V dc FHEAs to FPL dated February 15, 1990.

8-2

4s h

~ ~ I ~ ' '

~' h 1 ll ~ ~~

I1

' ' 11 ~

I I 1~

~ 11 11'1 1 1

~

I 1'

I I ~)

I

I

~

)I

~

~ 1'

~ I I ~1 ~, ~

~ I ~ 1 '1 )

1~

~ I 1' )

I) ' ) ~

)'

~

1'l

' )

1 I I )

I )

I

)

I)'

~

~

I 1

~

~I I\

~ '1 '1 I I

~ ~

' ~ ~ 11 '

1'l '

~ ~ li '

1 ~ ) ) ~

~

' ' 1' ~\

1

' ' 1 ~

I a'

graf

) eea ) eoa ) oooo Mea II'c w<<S<<oe s )is gs gs shelce esses w a 1$$ WC Iae <<>><<s

) cool )e

$1 WC

)~ )

) Ioel ) IS<<

<>. Iee <<$ I>>qa <<I SO. IOO <>of v>>>ea IICI ~

>>SI $

loose C~

ee feose $$ C<<$ $ $ a $<<$ 1$ ce Cseo$

NOTES:

L ef s ~~

<<~o CK ws<<es

<<s IK<<CK sewse

~eapcRI >>wl K owl $ $

ae K clcolc II

<<sees Kslss oeesee essssL S $ $$ SSII ISC ~ KC>>SIC KC SC s osss>> sssK oee>>IK Kelecec <<4 KVKMCK seeea DK <<>> IK as Ks7.

OCIISW se

$11

~

s

~

~

LEGEND Ot<<1$ % $$

WI<<$$ $

KI CK<<loess>>e<<esi KCIS>>

INDICATES EOUIP ADDED BY EPS ENHANCEMENT PROJEC ANO OTHER MOOS fLORIDA POWER dr LICTTT COMPANY

<<sea gK, II <<I TURKEY POINT.,PTAHI UNITS 0 4 e WI 4$ EMERCENCY POWER SYSTEM ENHANCEMENT DC ONE, llNE DIACRAM 1 14IIR1: 2

1~ J 5if r IC Cl