L-86-243, Requests Expedited Review of Encl Evaluation of Emergency Diesel Generator Loading for Concurrent Operation of Units 3 & 4,per 860402 Confirmatory Action Ltr & for Unit 4 Restart. Two Addl Items Will Be Evaluated Separately.Fee Paid
| ML17342A581 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 06/12/1986 |
| From: | Woody C FLORIDA POWER & LIGHT CO. |
| To: | Rubenstein L Office of Nuclear Reactor Regulation |
| References | |
| L-86-243, NUDOCS 8606170014 | |
| Download: ML17342A581 (63) | |
Text
REGULATE)(Y INFQRNATION DISTR IBUTIO mYBTE>'l (R IDB)
ACCESSION NBR: 8606170014 DGC. DATE: 86/06/12 NOTARIZED:
NQ FACIL: 50-250 Turl:eg Point Plant>
Unit 3>
Flov ida Pommer and Light C
50 251 Turl(eg Point Plant>
Unit 4>
F1ov i da Potoer and Light C
AUTH. NANE AUTHOR AFFILIATION NODDY C. G.
Flov ida Power Zc Light Co.
REC IP. NANE RECIPIENT AFF1LIATIGN RUBENSTEIN> L S.
PWR ProJect Directorate 2
DOCKET 0 05000250 05000251
SUBJECT:
Requests expedited v eviee of encl evaluation of emergency diesel generator loading for concurrent opev'ation of Units 3 8c 4> pev'60402 confirmatory action ltv 5 for Unit 4 v estav t.
Two addi items ei I 1 be evaluated separatelg.
Fee paid.
DISTRIBUTION CODE:
A001D CGPIEB RECEIVED: LTR ENCL SIZE:
TITLE:
QR Submittal:
Geneval Distribution NOTES:
RECIPIENT ID CODE/Nhl'IE PWR-A EB PWR-A FQB PNR-A PD2 PD 01 PNR-A PSB INTERNAL: ADN/LFND EQ FIL CGPIEB LTTR ENCL 1
1 5
5 1
1 1
0 1
1 1
1 RECIPIENT ID CODE/MANE PWR-A EICBB PWR-A PD2 LA Nc DONALD> D PWR-* RSB ELD/HDB4 NRR/GRAS RQN2 COP IEB LTTR ENCL 2
2 0
1 1
1 0
1 0
1 EXTERNAL: 24X LPDR NSI C 03 05 1
1 1
1 1
EQSQ BRUSQUE> S NRC PDR 02 1
1 1
1 TOTAL NUNBER GF COPIES REQUIRED:
LTTR 24 ENCL 20
0 a
b,l t
"I
FLORIDA POWER & LIGHTCOMPANY JUN I 2 1888 L-86-203 Office of Nuclear Reactor Regulation Attention:
Mr. Lester S. Rubenstein, Director PWR Project Directorate II2 Division of PWR Licensing - A U. S. Nuclear Regulatory Commission Washington, D.C.
20555
Dear Mr. Rubenstein:
Re: Turkey Point Units 3 and 0 Docket Nos. 50-250 and 50-251 Emer enc Diesel Generator Load Evaluation Attached is the evaluation of emergency diesel generator loading for concurrent operation of Turkey Point Units 3
and 0
required by the NRC Region II Confirmation of Action letter dated April 2, 1986.
Florida Power 8 Light Company requests your expedited review and approval of this evaluation in order to support the scheduled activities associated with the restart of Unit 0.
11 All plant modifications and procedure revisions necessary to support the attached evaluation will be completed prior to the Unit 0 restart.
Required operator training will be completed before an individual operator assumes shift duties for the restart.
In the evaluation, there are several references to the "hot shutdown" unit.
This refers to the unit which is not assumed to have the postulated accident.
Under the recent Technical Specification addition of modes of operation, the hot shutdown unit refers to the unit in the hot standby mode.
On pages 11 and 12 of the evaluation, there is a discussion of two design features that willbe modified. As stated in the evaluation, these items are being evaluated independently of the EDG load evaluation.
A 10 CFR 50.59 evaluation and 3ustification for Continued Operation (JCO) will be developed for any interim actions required prior to completion of any required design modifications.
The evaluation's restriction on Computer Room/Cable Spreading Room temperature is being further evaluated to establish additional operational margin if possible.
8 o ADOCN, o>000 170014 Bbob128 PDR ADO PD p
Ll:1 goo
(
PEOPLE... SERVING PEOPLE
Page two In accordance with 10 CFR 170.12(c), FPL Check No. 1313 for $ 150.00 is attached.
If you have any questions regarding this submittal, please call us.
Very truly yours, C. O. Woo Group V President Nuclea nergy.
COW/TCG:de Attachment cc:
Dr. 3. Nelson Grace, Region II, USNRC Harold F. Reis, Esquire P NS-LI-86-191
~
I
SAFETY EVALUATION TURKEY POINT UNITS 3 Bc 0 (PTPN)
EMERGENCY DIESEL GENERATOR LOADEVALUATION
4
'II
Page 2 of 31 SAFETY EVALUATION TURKEY POINT UNITS 3 Bc 0 (PTPN)
EMERGENCY DIESEL GENERATOR LOADEVALUATION INTRODUCTION
'n Emergency Diesel Generator (EDG) load evaluation, conducted in December 1985, indicated that EDG loads were at acceptable limits.
The study also indicated that EDG margins were not excessive (see Table 1), and that any future change in EDG loading requirements must be fully evaluated prior to implementation.
Subsequent to completion of this study four items were identified by Power Plant Engineering that would affect EDG loadings; namely, The TMI-related Emergency Operating Procedures (EOPs) would require the addition of a charging pump in the initial (1-30 minutes) EDG load interval, and running a second high head safety injection pump in the second (30 minutes - 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />) EDG load interval.
Inplant CCW and ICW testing indicated that pump kW loads would be higher than those assumed in December 1985.
Environmental qualification (EQ) evaluations per 10 CFR 50.09 assumed that the Emergency Containment Coolers (ECCs) auto-connect and are not secured at 30 minutes.
Load center transformer losses had not been accounted for in prior EDG load evaluations.
There is insufficient EDG margin available to accommodate the potential EDG loading levels that would result from incorporating these additional loads without some compensating actions.
EDG LOAD LIMITS The EDG kW load ratings are provided in Table 2.
The values provided in the column FSAR are those contained in the FSAR at Table 8.2-2.
These values are consistent with those furnished by the VDG supplier.
The diesel engine manufacturer's ratings for the engine are expressed in BHP in Table 2.
These BHP ratings were converted into an equivalent kW rating.
The ratings supplied by the EDG supplier and those derived from the engine manufacturer's published BHP data are in general agreement.
Table 2 indicates that EDG loadings greater than the 2000 hour0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> rating of 2850 kW are severely restricted as to the duration that these loads may persist on the machine.
For example, a modest increase of 100 kW above 2850 kW reduces the
Page 3 of 31 recommended running time dramatically (from 2000 to 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />).
Accordingly, a
load limit of 2850 kW was established for short-term continuous loadings.
Short-term is interpreted to mean a few hours.
During post-LOCA load management activities, loads must be added and removed from the EDGs.
The 168 hour0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br /> rating of 2950 kW was selected as a procedural limit to allow for transient load conditions that may occur during load management activities.
The current PTPN Technical Specifications at 0.8.l.c.8 require periodic testing and surveillance "Verifyingthat auto-connected loads to each diesel generator do not exceed 2750 kW."
This limit refers to the cold start loading of the EDG at the onset of the LOCA with a loss of offsite power (LOOP).
The Standard Technical Specifications (STS) for Westinghouse Pressurized Water Reactors at 0.8.1.1.2.f.8 state that auto-connected loads to each diesel generator are to be verified to ensure that. they "do not exceed the 2000 hour0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> rating...", which is 2850 kW for the Turkey Point EDGs.
The current Technical Specifications are more restrictive than the STS requirement.
Based on the above, the load limits on the EDG are established as follows:
Transient 2950 kW Short-Term Continuous 2850 Auto-Connect 2750 The 2850/2950 kW load management philosophy has been incorporated in the plant procedures by the incorporation of the following CAUTION or its equivalent.
(The markings in the caution statement are set based on instrument calibration.
The Red mark accounts for instrument error such that the 2950 kW limitis not exceeded.)
CAUTION Loading on the EDG should not exceed the Orange mark (2000 hour0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br />,rating of 2850 kW). However, loads placed on the energized 0 kV buses (the power source) may approach the Red mark (168 hour0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br /> rating of 2950 kW) on the wattmeter for short periods.
The PTPN EMD Model 20-605EO engine has been tested by the supplier at 2950 kW for about 30 minutes, and it is routinely tested to 2750 kW in accordance with PTPN Technical Specification 0.8.1.c.6.
, Additional testing of the EMD Model 20-605EO engines at loads from 2500 kW to 3050 kW has been conducted at other installations.
The results of a survey conducted by Power Plant Engineering is provided by Table 3. It indicates that two EMD, Model No.
20-605EO EDGs have been run at 2850 kW for 23.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> followed by a 1/2 hour run at 3050 kW.
I 0
k
Page 0 of 3l CORRECTIVE ACTIONS Corrective actions are required prior to two unit operation.
These actions allow the EDGs to accommodate the recently defined loading requirements.
Each item (and supporting basis) is discussed in detail below.
In summary, they are:
o Secure a RHR pump on the accident unit at 30 minutes.
o Disable the auto-connect feature of the Instrument Air Compressors (IACs).
o Disable the ability of the IACs to reload upon Safety Injection Signal (SIS) reset.
o Provide EDG independent IACs.
(Equipment to support IAC operation willalso be EDG independent.)
o Disable the ability of turbine-related loads to reload upon SIS reset on both units.
o Initiate the Computer Room/Cable Spreading Room (CR/CSR)
. function after one hour.
o Provide Computer Room temperature indication in the Control Room.
o Disable the auto-connect feature of the Normal Containment Coolers (NCCs) on the non-accident unit.
o Upgrade; as required, NCCs and support equipment for operation at
'elevated temperature.
o Secure a Containment Spray (CS) pump at 30 minutes and maintain two Emergency Containment Coolers (ECCs) running beyond 30 minutes.
o Disable the auto-connect of boric acid-related loads (except heat tracing) when two EDGs start and operate.
The plant procedures require that a second high head safety injection (HHSI) pump be run in the second EDG load interval.
To compensate for this, the residual heat removal (RHR) pump is secured in this load interval. This results in a break independent LOCA scenario; i.e., the operator response would be the same for all break sizes.
Analysis demonstrates that securing a RHR pump at 30 minutes does not affect
-the ECCS evaluation.'Securing the RHR pump also extends the duration of the injection phase of the post-LOCA period.
For the one EDG large break LOCA (LS LOCA) case, the injection phase is extended from about 39 minutes to 50 minutes.
Thus, more time is provided for the operator to stabilize the plants prior to entering recirculation.
Page 5 of 31 o
The newly implemented PTPN version of the TMI-related EOPs include provisions to run two HHSI pumps and no RHR pump for the injection phase beyond 30 minutes.
They also provide for running one HHSI pump and one RHR pump in the recirculation phase.
Upon SIS reset, the lACs and certain turbine-related loads would be re-enabled.
The design has been altered to prevent these loads from auto-connecting to the EDGs upon SIS reset.
EDG independent IACs will be provided to ensure the availability of instrument air post-LOCA.
Equipment to support IAC operation willalso be EDG independent.
Furthermore, the plant procedures willnot allow the manual loading of turbine auxiliaries unless EDG loading permits their addition.
Sufficient EDG capacity is not available to power turbine-related loads when only one EDG is operating.
o
. The plant procedures willinstruct the operator not to load turbine loads for the accident and non-accident units when only one EDG is available during a LOCA concurrent with a loss of offsite power (LOOP).
Inplant testing has demonstrated that the Computer Room/Cable Spreading Room (CR/CSR) AC function is not required immediately after the occurrence of a LOCA concurrent with a LOOP.
The function will not be required for one hour.
Computer Room temperature indication will be installed to allow the operator to monitor the temperature.
The CR/CSR AC function willbe initiated from the Control Room one hour post-LOCA to ensure that the 90oF ambient'emperature limit for the Safety Assessment System (SAS) is maintained.
The Inadequate Core Cooling System (ICCS) ambient temperature limit is 100oF.
The Computer Room is maintained at or below 67oF during normal plant operation.
The plant procedures willinstruct the operator to start the CR/CSR AC one hour after accident initiation to assure SAS availability.
The plant procedures will instruct the operator to maintain the Computer Room at or below 67oF during normal operation to assure SAS availability.
Figure 1 provides the containment temperature response on the hot shutdown (non-accident) unit as a function of the number of NCCs operating.
It indicates that if one NCC is operated, the containment temperature willrise from 120oF to a value asymptotic to 133o at more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.
Figure 2 indicates that the one NCC need not be started for the first hour post-LOCA. At one hour, when the NCC is
- started, the temperature is,.l27oF.
Afterwards, the temperature profile conforms to that shown on FigUre l. Evaluation of the unit in the hot shutdown condition indicates that operation at elevated temperature is acceptable.
The NCC-related component evaluation assumed that NCC-related items on the unit currently in operation were identical to those on the unit cuirently in cold shutdown.
Jf II
Page 6 of 31 o
A documented field verification will confirm the validity of NCC-related-component assumptions.
If required, modifications will be made prior to operation to ensure continued NCC operability at the elevated temperature.
The NCC-related component evaluation also indicated that the NCC damper motors are designed for operation in an ambient temperature of 130oF.
This temperature will be reached in about 27 hours3.125e-4 days <br />0.0075 hours <br />4.464286e-5 weeks <br />1.02735e-5 months <br /> post-LOCA. If a second NCC is not in operation within the first 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> or so, the damper function will be disabled manually to preclude spurious damper operation.
o The plant procedures will instruct the operator to disable the NCC damper function at 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> when one NCC is run continuously.
Starting two 'CCs per unit with a LOOP and no LOCA will maintain containment temperatures on the hot shutdown unit(s) below 120oF.
o A design modification will be implemented to prevent EDG auto-loading of the non-accident unit's NCCs.
o The plant procedures will instruct the operator to load one NCC on the EDG at one hour post-LOCA. Actuation of the NCC will be from the Control Room.
o The plant procedures willplace the following limitations on operation of the non-accident unit.
This will enable the operator to ensure stable hot shutdown conditions are maintained:
Charging flow will be established during the 1-30 minute EDG load interval.
Pressurizer heaters (150 kW) will be established in the 30 minute-one hour EDG load interval.
The MSIVs and Main Feedwater Isolation Valves will be closed immediately post-trip to minimize Steam Generator heat loss.
After restoring and stabilizing Steam Generator level, AFW flow to the Steam Generators will be minimized to maintain RCS Tavg above 531oF Pressurizer pressure willbe maintained at or above 1800 psig.
Containment pressure willbe maintained below 3 psig.
Steam Generator pressures willbe maintained equal (i.e., within 50 psid).
Steam Generator pressure willbe maintained above 600 psig.
o The plant procedures willinstruct the operator to start two NCCs on each unit immediately after a LOOP without safety injection, i.e.,
Page 7 of 3l Figure 3 shows the effect on the containment temperature transient if the CS pump is secured after 30 minutes.
The containment pressure transient exhibits an equivalent response.
Evaluation of these transients against the EQ envelope confirms that the CS pump may be secured at 30 minutes, provided that two ECCs are maintained in operation thereafter.
o The plant procedures will allow the operator to secure the operating CS pump at 30 minutes for the one EDG operating post-l.OCA
- scenario, and willinstruct the operator to maintain two running ECCs beyond 30 minutes.
The fan horsepower required for operation of an ECC or an Emergency Containment Filter (ECF) is proportional to the containment
- pressure, which varies with time.
The fan horsepower was calculated based on the peak pressure occurring in the first and second EDG load intervals.
The calculation assumed the CS pump was secured at 30 minutes.
Boric acid-related loads (boric acid transfer pumps and boric acid tank heaters) auto-connect when two EDGs start and operate.
These loads are not required on-the accident or hot shutdown unit and do not auto-connect when only one EDG is available.
This auto-connect feature willbe disabled.
o The plant procedures will instruct the operator, that the. boric acid tank heaters may remain deenergized for up to eight hours without solidification of the solution therein.
Inplant testing and analysis confirmed that, as stated in the FSAR at Subsection 9.9.l, only one Control Room Air Conditioning (CRAG) unit is required.
Two CRAG units are auto-connected to the EDG.
o The plant procedures will instruct the. operator that one of the two CRAC units may be secured from the Control Room to facilitate EDG load management activities.
PUMP kW VALUES The pump kW values for the 1CW and CCW pumps were determined by inplant testing.
The testing was conducted at design flow values and beyond design flow values.
This provided inplant test data over the range of possible pump conditions, and reconfirmed the validity of plant startup test data.
Figures 0 and 5 provide a plot of kW versus flow for CCW and ICW pumps, respectively.
Table 0 provides pump kW values and the confidence, level associated with the value.
Table 5 compares the pump kW values used fn.the FSAR and this load evaluation.
The sum of individual pump kW values utilized in the current analysis is about equal to the sum of the FSAR values.
The computer code utilized for the failure modes and effects analysis (FMEA) required that individual pump kW values be utilized as an input for the FMEA.
Limiting cases identified by the FMEA were analyzed utilizing the 9596 confidence value for "total pump kW." Table 6 provides the expected total pump kW and the 9596 confidence level pump kW for the one and two EDG operating cases.
Page 8 of 31 FSAR LOADEVALUATIONCOMPARISON pump kW comparisons between the FSAR and current load evaluations are discussed in the preceding section.
Tables 7 and 8 provide a comparison with other assumptions for the FSAR and current case.
The principal differences are:
o The NCCs are secured for one hour in the current evaluation, as'are the CR/CSR AC units.
o Charging pump capability is provided earlier in the current evaluation, with pressurizer heaters provided after 30 minutes.
o The IACs and turbine-related loads are disabled from reloading on SIS reset.
o The RHR pump is secured at 30 minutes in the current evaluation.
o The CS pump is secured at 30 minutes in the current evaluation.
o Two HHSI pumps run during the entire injection phase in the current evaluation.
o Two ECCs are run beyond 30 minutes in the current evaluation.
Differences between the FSAR and current EDG load evaluations are discussed in the paragraphs that follow.
Tables 7 and 8 provide a comparison of EDG kW loads, based on the sum of individual pump kW values, with those contained in the FSAR.
The EDG load evaluation sum must be corrected to account for a total pump kW value with a 9996 confidence level, as shown.
Thus, the current load evaluation results in EDG kW loads less than the FSAR.'he FSAR total kW estimate remains bounding.
Securing the NCCs on the hot shutdown unit raised the containment temperature.
The ability to maintain hot shutdown conditions is not adversely affected by the temperature increase.
For the case of LOOP only, the operator starts two NCCs and the temperature is reduced to less than 120oF for non-LOCA LOOP scenar ios.
Providing charging capability and pressurizer heaters on the hot shutdown unit earlier in the post-trip period than assumed in the FSAR enhances the operator's ability to maintain hot shutdown conditions.
Disabling the reload capability upon SIS reset for the.IAC and turbine-related loads reduces currently required operator actions prior to SIS reset.
Starting one NCC at about one hour from the Control Room does not impose any undue burden upon the operator, nor does the starting of a CR/CSR AC from the Control Room.
Disabling the NCC damper function occurs long term.
The accident and non-accident units would be stabilized on the order of an hour or so.
This activity would take place at about 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.
8
Page 9 of 31 For a small break LOCA (SB LOCA) the operator secures a RHR pump during injection.
The FSAR evaluation would require the operator to secure one HHSI pump at 30 minutes.
These two different operator actions would be replaced by one action; namely, secure a RHR pump at 30 minutes for all LOCAs.
Thus the required operator actions are reduced by this break independent LOCA response scenario.
The operator no longer has to differentiate between a SB LOCA. and a LB LOCA since his actions are identical for all LOCAs.
In
- balance, implementation of the assumptions associated with this load evaluation reduces the operator actions currently in place.
Securing one safety injection pump at 30 minutes is consistent with the FSAR load evaluation.
Securing the CS pump at 30 minutes does not invalidate equipment qualification.
Running the ECCs beyond 30 minutes is consistent with current EQ requirements.
Running two HHSI pumps for the injection phase beyond 30 minutes, and running one HHSI and one RHR pump in the recirculation phase is consistent with the newly implemented TMI-related EOPs.
The comparison with FSAR requirements indicates that implementation of the current load evaluation does not involve an unreviewed safety question since the FSAR load evaluation bounds this case.
Individual 50.59 evaluations will be provided with each plant design change
- package, and where appropriate for procedural changes.
FAILURE MODES AND EFFECTS ANALYSIS(FMEA)
A FMEA was conducted to evaluate the ability of the EDGs to accommodate single failures in the electrical system for various combinations of accident assumptions.
Potential single failures were evaluated by Power Plant
'ngineering and those resulting in the highest EDG loadings were included in the F MEA.
The results of the FMEA are provided by Table 9. Limiting cases defined by the FMEA are reviewed to determine the appropriate EDG loading conditions, since FMEA assumptions are overly conservative.
With no single failures, the maximum auto-connected load determined by the FMEA calculation is 2673 kW.
For the one EDG cases, the maximum auto-connected load is 2690, determined by the FMEA calculation kW. Thus, for those cases that are subject to the Technical Specification auto-connected limit of 2750 kW, all FMEA cases are within this limit..
Alltwo EDG cases assume that each EDG auto-connect's,the following pumps:
2ll 2
t
Page 10 of 31 In the 1 - 30 minute EDG load interval, one HHSI is assumed to be secured by the operator on each EDG.
The operator has considerable load management flexibility when two EDGs are operating bec'ause the number of large running pumps is greater than required.
On the other hand, the operator has little load management flexibilitywhen one EDG is operating.
The single EDG case is more restrictive from a
Iqad management standpoint.
The FMEA e'valuated the case where a LOCA occurs with one unit at cold shutdown, even though the FSAR design basis does not address this case.
The analysis assumed that the RHR on the cold shutdown unit is not restarted until 30 minutes after the LOCA.
This assumption is acceptable without further restriction if the LOCA occurs in excess of 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> after shutdown.
The probability of a LOCA not occurring during this 30 hour3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> period is acceptably low.
For example, assuming one refueling per unit per year, the probability is about 6.8 x 10-3 less than that associated with the design basis event.
From an EDG loading standpoint, starting a RHR pump in the 1 - 30 minute EDG load interval is acceptable.
The EDG kW load based on a 9596 confidence level for total pump kW is 2809 kW, and the load based on expected pump kW is 2826 kW.
One CRAG unit (27 kW) can be secured prior to starting the RHR pump on the cold shutdown unit.
The plant procedures will instruct the operator not to initiate RHR on the cold shutdown unit for the first 1/2 hour when a LOCA occurs with one unit in cold shutdown and only one EDG is operating.
RHR on the cold shutdown unit would be started after RHR or CS is secured on the accident unit.
If the LOCA. on the running unit occurs within 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> after shutdown, the operator would be instructed to stop one of the two running CRAC units and start the RHR pump during the first 1/2 hour.
If the LOCA occurs when the Reactor Coolant System is drained to the mid-nozzle height, the operator shall restore RCS level prior to restart of RHR.
Two charging pumps with suction from the RWST provides sufficient flow to restore RCS level.
The single tie breakers connecting load centers with A and B power supplies are maintained open with either unit at power.
This precludes any unacceptable interaction between the A and B power supplies.
o The plant procedures will ensure that load center tie breakers are racked out for both the A R B trains when either unit is at power, unless a safety evaluation or Technical Specification limitis provided to justify short-term operation with the breaker closed.
Page 11 of 31 For the electrical system single failure cases, only the MCC vital to non-vital tie breaker cases result in a high auto-connected EDG load on one of the two operating EDGs.
This loading can be mitigated immediately by securing one or two large pumps on the heavily loaded EDG.
The MCC D vital to non-vital evaluation assumes that the Battery Room AC (22kW), two Control Room ACs (50 kW), and all boric acid heat tracing (20 kW on EDG B) are fully loaded, on the EDG during the 0-1 minute auto-sequencing of loads.
If rione of these loads were demanded by process controls during this one minute interval, then the calculated load would be reduced by 96 kW. A further reduction in kW is obtained by utilizing the most probable value for total pump kW, i.e., the expected value provided by Table 6.
Adjusting the FMEA results indicates that the expected load on EDG B, due to a MCC D tie breaker failure, would be anywhere from 2805 to 2901 kW.
A similar evaluation for the other MCCs indicates that the EDG A load due to an MCC 3A tie breaker failure could be anywhere from 2805 to 2852 kW, and for MCC OA from 2836 to 2883 kW.
The expected auto-connect kW load is within the EDGs'950 kW limit; is greater than the Technical Specification auto-connect limit of 2750 kW; and is in the order of the STS requirement of 2850 kW. Thus the single failure of a MCC tie breaker results in an unanalyzed PTPN EDG auto-connect loading scenario on one of the two operating EDGs.
This scenario approaches current NRC requirements embodied in the STS.
The FMEA vital to non-vital tie breaker failure results assume that Main Transformer cooling equipment, which is normally supplied from MCCs 3A and OA, is powered from the alternate C-Bus supply.
Thus, until design changes are implemented:
o The plant procedures shall be modified to ensure that the Main Transformer cooling equipment normal supply breakers on MCCs 3A and OA are open, and that the alternate C-Bus supply breakers are closed.
The FMEA identified two design features that willbe modified; viz:
o MCC OA receives its normal power supply from LC OA. Auxiliaries for EDG B are powered from this bus.
o As shown in FSAR Figure 8.2-2, a single breaker connects the vital to non-vital bus sections of MCCs D, 3A and OA. Failure of one of these breakers to open'ill cause both the vital and non-vital bus section loads to be auto-connected to the EDGs.
The effect of these design items are being evaluated independently of the EDG load evaluation.
Appropriate design changes will be developed to modify these items such that they do not have a potential for adversely affecting the ability
0
Page 12 of 31 of the EDGs to-accommodate required loads.
Appropriate safety, 10 CFR 50.59 and 10 CFR 21 evaluations will be conducted in parallel with this design effort.
A 50.59 evaluation and 3ustification for Continued Operation (3CO) will be developed for any interim actions required prior to completion of any required design modifications.
DURATIONOF ONE EDG OPERATION: PLACING THE PROBLEM IN PERSPECTIVE The occurrence of a LOCA followed by a unit trip, which is postulated to result in a loss of offsite power to both units, and then a subsequent failure of one EDG, constitutes the design basis accident scenario.
This is an unlikely event.
The probability of occurrence of this event has been calculated, and is 2.3 x 10 -5 events/year.
Given that the event has occurred, the AC power question is reduced to how long one EDG operation will be required.
To respond to this query, the LOOP events to FPL fossil and nuclear facilities since 1973, and the repair time associated with EDG failures for both the St.
Lucie and Turkey Point nuclear facilities have been analyzed.
The results indicate:
Number of Occurrences Median Time to Restore Mean Time to Restore EDG FAILURES 200 min.
600 min.
LOOP 27 20 min.
26 min.
The probability of not restoring AC power is dominated initially by the restoration of offsite power.
There are five Black Start Diesels installed at PTPN.
These diesels can also be used to power either the A or B buses via the C bus.
This power source can be made available within about 15 minutes after the decision is made to initiate this AC source.
It is thus reasonable to assume that this power source can be made available at about one hour post-LOCA.
The Black Start Diesel reliability analysis assumes that these units are well maintained and available for nuclear plant use.
Maintenance of these units will become the responsibility of the Nuclear Energy Department, and an appropriate maintenance program'ill be implemented to ensure their continued availability.
Figure 6 provides the results of the evaluation of the failure to restore an additional source of AC power.
Assuming restoration of offsite power and EDG repair only, the probability of not restoring one of these sources decreases by an order of magnitude per hour.
That is, at one hour, failure to restore an additional AC power source will occur 1 out of 10 times,.at two hours 1 out of 100 times, at three hours 1 out of 1000 times, etc.
If the Black Start Diesels are started, 1 out of almost 1000 times the attempt to
ll
Page 13 of 3l power an A or B-bus from the diesels will not be successful.
Thus the presence of the Black Start Diesels dramatically reduces the probability of not restoring an additional AC power source.
It is reasonable to conclude that an additional AC power source can be made available on the order of an hour post-LOCA.
CONCLVSIONS The V.DG loading evaluation, and associated plant changes and changes to the plant procedures, result in a substantial improvement in EDG margin when compared to the FSAR (see Tables 7
dc 8).
In addition, the reliance on operator action will be reduced after correction of the MCC tie breaker and EDG B
auxiliary power supply design items are implemented.
Implementation of the evaluation's assumptions demonstrates that the FSAR load evaluation remains bounding, and that the EDGs meet all loading requirements set forth in the PTPN Technical Specifications.
Detailed evaluations and supporting calculations are available at FPL for review by NRC.
Each PC/M will contain a written safety evaluation and 50.59 evaluation.
Page 10 of 31 TABLE 1, EDG k% LOADINGMARGINBASED ON 12/85 EVALUATION Auto-Connect 1 min. 30 min.
30 min. - 1 hr.
EDG Loading+
Margin to 2950 kW EOP Limit Margin to 3050 kW Yi Hr. Rating Margin to 2750 kW Tech. Spec.
Auto-Conn. Limit 2707 kw NA 3
(0.1%)
2875 75 (2.5')
175 (5.7X) 2790 156 (5.3%)
256 (s.w)
+ From 12/85 EDG Load Evaluation Study
4
Page 15 of 31 TABLE 2 EMERGENCY DIESEL GENERATOR RATINGS BHP>>
kW~+
FSAR kW Base Continuous Rating 3600 2567 2500 110% Rating (2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />s/20 hours) 3960 2828 2750 2000 Hour/Year 200 Hour/Year 168 Hour/Year 0 Hour/Year 1/2 Hour/Year 3950 0100 0150
'0225 2821 2930 2966 3020 2850 2950 3050
'+ 'iesel engine manufacturer's rating for EMD Model No. 20-605EO engine.
++ Calculated assuming a 60 horsepower fan and 0.972 generator e fficiency.
Page 16'of 31 TABLE 3 TESTING OF EMD MODEL 20-605EO EDGs NO. EDGs TEST TIME 22 Hours 2 Hours 22 Hours 2 Hours 0 Hours 2 Hours 5 Minutes+
6 Minutes+
20 Hours 23.5 Hours 0.5 Hours POWER LEVELkW 2068++
2900 2500 2900 2600 2850 2950 3050 2750 2850 3050 Run until stabilized at load.
'+
Average value, load varied frown 2020 to 2520 kW.
'h
Page 17 of 31 TABLE 0 PUMP k%'ALUES AND CONFIDENCE BOUNDS PUMP HHSI HHSI RHR RHR CS CS ICW ICW CCW CCW SEDGs+
PUMP kW 302 302 220 195 216 216 265 265
. 380
. 365 CONFIDENCE (%)
95 95 95 95 95 95 99
'99 95
+ With two EDGs operating the flow per pump may be reduced, which results in a kW per pump reduction.
lf
Page I8 of 31 TABLE 5 PUMP kW COMPARISON FEAR 6/86 RHR CS HHSI CCW ICW 220 199 612 760 520 2323 220 216 600
~ 760 530 2330
Page 19 of 31 TABLE 6 TOTAL PUMP kW One EDG Case Less than 30 Minutes
~lt Pum s
Ex ected Pum kW HHSI RHR CS CCW ICW 95% Total kW = 2278 585 210 209 701 507
~
2256 One EDG Case Greater than 30 Minutes
~// Pum s
Ex ected Pum kW HHSI RHR CS CCW ICW 2
0 0
2 2
585 0
0 701 507 1833 95% Total kW = 1850
I ~
Page 20 of 31 TABLE6 TOTAL PUMP kW Two (2) EDG Case Less than 30 Minutes
// Pumps Per EDG Ex ected Pum k W HHSI RHR CS CCW ICW 585 189 209 712 507 2202 959o Total kW = 2220 One EDG Case Less then 30 Minutes (One Unit in Cold Shutdown)
// Pumps Per EDG Ex ected Fum kW HHSI
2 1
2 2
585 028 209 701 507 2070 95% Total k W = 2093
Page 21 of 31 TABLE 7 ONE EDG AVAILABLE COMPAIUSON WITHFSAR k% LOADS 0 - 30 MINUTES HHSI RHR CS CCW (Accident Unit)
CCW (Hot Shutdown Unit)
ICW (Accident Unit)
ICW (Hot Shutdown Unit)
Normal Containment Coolers Emergency Containment Coolers Emergency Containment Filters Battery Chargers Charging Pump Pressurizer Heaters Turbine Loads Emergency Lighting Control Room AC BA Heat Tracing EDG Auxiliaries Miscellaneous Loads Load Center Transformer Losses Battery Room AC H2 Analyzer Related Security Building Transformer Computer Room/Cable Sprdg. Room Boric Acid Pump 9596 Total Pump kW Correction FSAR TABLE 8.2-2 612 220 199 399 365 262 262 120 50 122 0
0 0
0 00 00 00 6
22 0
AC 0
2771 EDG LOAD~
EVALUATION 600 220 216 380 380 265 265 0
00 100 0
101 0
0 31 50 00 17 9
10 22 10 8
0 0
2788
-50 2738 PTPN 3 dc 0 at P'ower, SIS on PTPN 3, EDG A fails (see Table 9).
Page 22 of 31 TABLE S ONE EDG AVAILABLE COMPARISON WITH FSAR kW LOADS BEYOND 30 MINUTES FSAR TABLE 8.2-2 EDG LOAD EVALUATION+
HHSI RHR CS CCW (Accident Unit)
CCW (Hot Shutdown Unit)
ICW (Accident Unit)
ICW (Hot Shutdown Unit)
Normal Containment Coolers Emergency Containment Coolers Emergency Containment Filters Battery Chargers Charging Pump Pressurizer Heaters Turbine Loads Emergency Lighting Control Room AC BA Heat Tracing EDG Auxiliaries Miscellaneous Loads Load Center Transformer Losses Battery Room AC
,. H2 Analyzer Related Boric Acid Pump Security Building Transformer Computer Room/Cable Sprdg. Room 95% Total Pump kW Correction AC 306 220 199 399 365 262 262 120 0
122 130 125 0
50 00 00 00 10 27 2737 600 0
0 380 380 265 265 0
30 70 190 0%%
150++
0 31 50 00 8
9 13 22 15 0
8 0
2502
-00 2502 PTPN 3 R 0 at Power, SIS on PTPN 3, EDG A fails (see Table 9).
++
Alternate charging pump and pressurizer heaters (150 kW) operation.
t f
Page 23 of 31 TABLE 9 (Sheet 1 of 3)
SUMMARY
No Sin le Failures EDG Auto-Conn.
1-30 Minutes 30 Min. I Hr.
o PTPN 3 R 0 at power, SIS on PTPN 3 o
PTPN 3 2 0 at Power, SIS on PTPN 0 o
PTPN 0 in Cold Shutdown, SIS on PTPN 3 One EDG Available A
B 2515 2673 2515 2593 2515 2673 23ll 2362 2311 2287 2029 2362 2330 2317 2330 2265 2399 2317 o
PTPN 3 R 0 at power, SIS on PTPN 3, EDG A fails to load o
PTPN 3 R 0 at power, SIS on PTPN 3, EDG B fails to load o
PTPN 0 in Cold Shutdown, SIS on PTPN 3, EDG A fails to load 2690 2693 0
2690 0
2788 2787 0
0 2686 0
2502 2501 0
0 2611
v rl
Page 20 of 31 TABLE 9 (Sheet 2 of 3)
SUMMARY
EDG Auto-Conn.
1-30 Minutes 30 Min.- 1 Hr.
o PTPN 3 R 0 at power, SIS on PTPN 3
,Battery 3B failure, breaker 3AB20 fails to close o
PTPN 3 R 0 at Power., SIS on PTPN 0 Battery 0A failure, breaker 0AA20 fails to close 2753 966 1010 2708 2099 1068 1116 2050 2005 1117 1161
,2005 1/2 EDGs Available
)
o
. PTPN 3 R 0 at power, SIS on PTPN 0, breaker 0AB21 fails to close o
PTPN 3 dc 0 at power, SIS on PTPN 0, breaker 3AB20 fails to close o
PTPN 3 R 0 at power, SIS on PTPN 0, breaker 3AA20 fails to close A
B A
B 2531 1169 2672 1056 1501 2653 2267 1360 2068 1052 1097 2008 2237 1391 2087 1030 1071 2072
0 r
Page 25 of 31 TABLE 9 (Sheet 3 of 3)
SUMMARY
Nonvital Tie Breaker Failures EDG Auto-Conn.
1-30 Minutes 30 Min. - I Hr.
o PTPN 3 R 0 in power operation, SIS on PTPN 3, MCC D tie breaker fails to open o
PTPN 3 4 0 in power operation, SIS on PTPN 3, MCC 3A tie breaker fails to open o
PTPN 3 2 0 in power operation, SIS on PTPN 3, MCC 0A tie breaker'ails to open 2515 2975 2926 2673
'2957 2673 2311 2665 2620 2060 2651 2060 2330 2619 2590 2068 2621 2068 MCC Transfer td,Alternate Su I
o PTPN 3 2 0 in power operation, SIS on'PTPN 3, transfer of MCC 0A to LC 0D o
PTPN 3 R 0 in power operation, SIS on PTPN 3, transfer of MCC 3A to LC 3D o
PTPN 3 8 0 in power operation, SIS on PTPN 3, transfer of MCC D to LC 0C A
B 2096 2692 2069 2718 2736 2051 2292 2381 2265 2008 2026 2207 2315 2332 2288 2359 2372
.2270
180 TEMPERATuRE (Q TURKEY POINT LONGTERM CONTAINMENTHEAllJP Page 26 of 31 170 160 NO NCC~s 150 140 130
> NGC QO 2 NCCs 110 0
10 20 30 40 50 60 70
&0.
TIME (HOURS)
FIGURE I
TURKEY POINT CONTAINMENTHEATUP 1 NCC STARTING AT 0 HR, 0.5 HR, 1 HR muPEkATuRE (F)
Page 27 of 31 NO NGGs 1HR
)26 124
.0.5 HR 0HR
%0 0
4 6
TIME (HOURS)
FIGURE 2 10
CONTAINMENT TEMPERATURE ANALYSIS NO CS ARKR 30 NMUKS Page 28 of 31 0
E N
u s
2 8
W 5 0 7 I 9 101830t05060728090100 200 300 Ce laC (oeconda+1
@+02) '
l4HA ORDINAL o
noes ~
aors FIGVRE 3
COMPONENT COOUNG WATER PUMP Page 29 of 31
-400 PUMP INPUT (KW) 350 300 0
~
L
~ 0 1 Po
~ ~
~ ~
0 l~
l ~
~ '
1 0
~
~
~ ~
~
~
~
~
~
~
P V
~
uLPPPPPL 4
~
~
P
~
~
~
~~
~
~
~
~
~
~
0 ~ l 0
. ~"t.""f"Pff.--".
~
f 0
\\
~
~
~
~
~
~ ~ f L
4 00 L
0 4
~
~
l ~
0 1st "-.r.-"Lh-l L
1 f
f
~
~
P
~-f."
~ L ~
~ ~
"r"
~
~
~ ~
~
~
~l ~
~ ~
~
~
~ ~
~
~
~
l r
~
~ l 0
~
~
0 ~
~ ~
0 (00 J
".).'-.4" ~
00 0L 4
0
~ P 0
~
~.t F 0'
~
~
0 r
~
~ l 00
~
~
~
~
~
l ~
4 ~ ~
I ~
L
~ ~
~I
~
J
~
0 P
0 0
~
0 0 ~
~
~ 0 0 ~
~ 0
~
~
~ ~
~ ~
+
~
~
~
Y
~
~ ~
~
~
'l
~
~.
J
~
~
fi
'l
~
~
~ ~ L
~
f0
~
L
~
~ f
~ l ~ ~
~
~
~ ~
~ ~
f
~,l
~
0 1
~
~
~
4
~
~
~
J
~
J
~
~
0L P
0 L
~
J
~
0L
~
Joo
~
~
~
~
0
~
~ 1 ~ 010
~
~
P 00
~ ~
~
~
~
4
~
~
~ ~ 01 4
~
0 ~ 0 0
P
~
~
250 200 l ~
1 1 ~ Of
~
~ l 0
1 00
~
~
~
~
0
~
~
f 0
f 0 ~
0 0L
~
J
~
~ ~
1
~
~
~
of L
~Ii 0
~
~
~
'V
~
J
~
~
J 0
110 ~
~
J
~ ~
~ 0 ~
ot
~
0L
~ ~ 4
~ ~
0
~
~
~
~
~
~ L
~
~
~
~ P L
~
~
~
~
~
\\1
~
~ ~ r goo
~
oor ~ 11 ~ Po ~\\ ~ l~
f ~ 00 ~
~
~
~
~
~
~ ~ J
~
~
000L ii 40000400
~ ~
4
~ ~
~ l ~
~
r
~ ~
L
~
~
~
~
~ ~ 0 ~ 0 ~ 00 ~
0 ~ 10 V
4
~'
~
4
~
~
~
r ~
~ ~
~
~
~
~
~ 00L0 ~
~ J 0 ~
~ 0 L
~
~ olo ~ 0 ~
4
~
P
~ 00
~ 000
~
~ 0 i~
~
~
~
~
~
~
~
~
~
L
~
0
~ 0
~
~
040 ~
L 0 ~ 0 ~ ~
'V
~
~
~
~
J 00 fo
$ 00
~
~
~
~
~
~ ~ l 4
f" - t ~
~ h
~
~ of soi ~
EXPECTED 95K C~ONR ~EHC 0
STARTUP DATA
. ~
1Qb4 DATA 0
2000 4000 6000 8000
$0000 12000 NOOO 16000 18000 20000 FLOW (GPM)
FIGURE 0
INTAKE COOLING WATER PUMP Page 30 of 3l P
280 u~P iNPUT (eV) a r
o o ~
~
~
~
~
h 4
~
~
~
4
~
~
Oo
~
~
260 240 220 200
~ o
~t o
o) o o{
oh oo J
o oL
~ @o@
~to
~ o
~r
~ op
~ ot
~ ~
~ f
~
~
~ o ~
~
~
~
~
o t
0 o
~
~r
~ oo ~ >>oto ~i oo ~
g O
~
~
~
~
~
~
~
~
~
~
~
~
~
f ~--.)..-.{.""f
~ ~".)-r" -.f-
~.
ot
~
i-~-f~ N--
V
~
~ M
~
~
~
lo o Loooo4ooo
~
~ ~ ~
L o4
~
oo ~
~
o o4
~
~
~
~
~
~
~ r
~
4
~
~
~
~
~
o>
~
~
~
~
ooot ooV
~
~
to
~
~ o
~
~
~
r
~
~
~
~
~
~ ro
~ o
~
j ~
~ t
~ )
~ f
~
o oooo+
~
f~ o ~ o too
~ oo+
~
~ )
~
oh
~ {
oJ
~
o ~ o ~ or o ~
+o o o 0 o
~ ~
~
~
~ r
~ oh J
~
~
~
~
~
r
~
~
ro
~
~ ooot r o
~
~
~
~
~
~
~
~
~ ~
I
~
~
~
~
~
oo L o 4o 4
~
o 4 oo
~ oooo Lo
~
~ ooo
~ o o o
~
~
~
~
~
)oooo
~)
~ a
~
~..ioooo..o)
~ \\
~
~
oooo{
~ ofo
~
180.
160 140 o
~
~L
~t
~Lo
~ o
~
~
~
~
~
~
~L J
~
~
~
'.l--
~ ~
~ Jo ~
~ o t
~ L ~
oo J oo
~ oooooL
~
~
~
> ~
oh ~ ~
J
~
~
~5
~ 4
~ ~ 0 ~ o
~
t
~ o L
~ o o+ 0 ~ o ~ Q
~oro
~ oLo ot
'lo
~
0
~Lo J
~ oo ~ ot
~
oo)o o o+o ~ oo ~ o Q ooo ~
J oo ooor oo
~
o l
~
o t ~
L ~
~
~
~ o
~ ~
~ t h
o
}oooo o ~ ofooooot ~
~ ~
~ {
~ o}
o EXPECTED QSX COHflDENCE QQX CONfiDEHCE 0
STARTUP
~ 'IQId DATA 2000 4000 6000 8000 10000 12000 NOOO 16000 18000 20000 FLOW (GPM) 1, FIGURE 5
0 i
FAILURE TO RESTORE ADDITIONALAC POWER PROBABIUTY Page 31 of 31 0.$
0.01 0.001 0.0001 0.00001 0.000001 0.000000$
L 0.00000001 0
TIME (HOURS)
FIGURE 6
0