IR 05000528/1994201
| ML17310B226 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 04/06/1994 |
| From: | Imbro E, Koltay P, Roy Mathew Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML17310B225 | List: |
| References | |
| 50-528-94-201, 50-529-94-201, 50-530-94-201, NUDOCS 9404200052 | |
| Download: ML17310B226 (34) | |
Text
U.S.
NUCLEAR REGULATORY COHHISSION OFFICE OF NUCLEAR REACTOR REGULATION NRC Inspection Report:
50-528/94-201, 50-529/94-201, and 50-530/94-201 License Nos.:
NPF-41, NPF-51, and NPF-74 Docket Nos.:
50-528/529/530 Licensee:
Arizona Public Service Company Facility Name:
Palo Verde Nuclear Generating Station, Units 1, 2 and
Inspection at:
Palo Verde Nuclear Generating Station Inspection Conducted:
February 28 through Harch 4, 1994 Inspection Team."
NRC Consultant:
Roy K. Hathew, Team Leader, NRR Amar Pal, NRR David Shum, NRR Frank Gee, RIV Omar Hazzoni, Parameter Prepared by:
Roy Hathew, Team Leader Team Inspection Development Section B
Special Inspection Branch Division of Reactor Inspection and Licensee Performance Office of Nuclear Reactor Regulation
'r -6-~i "
Date Reviewed by:
-'8 Peter S. Koltay, Section Chief
~
Team Inspection Development Section B
Special Inspection Branch Division of Reactor Inspection and Licensee Performance Office of Nuclear Reactor Regulation Da e
Approved by:
Eugene V. Imbro, Chief Special InspectIon Branch Division of Reactor Inspection and Licensee Performance Office of Nuclear Reactor Regulation Date
I l
l
)
EXECUTIVE SUMMARY.
TABLE OF CONTENTS Pa<ac 1.0 INSPECTION BACKGROUND AND OBJECTIVE 2.0 COPING DURATION ANALYSIS 3.0 STATION BLACKOUT COPING SYSTEMS 3.1 Condensate Inventory 3.2 Station Battery Capacity 3.3 Auxiliary Feedwater and Steam Relief 3.4 Compressed Air 3.5 Effects of Loss of Ventilation 3.6 Containment Isolation Valves 3.7 Reactor Coolant Inventory
.
3.8 Emergency Lighting and Communications 4 '
PROCEDURES AND TRAINING 4. 1 Severe Weather and Restoration'rocedures 4.2 Emergency Operating Procedures 4.3 Reliability Program
.
4.3.
.
4.3.2 Alternate AC Source 4.4 Station Blackout Training
.
5. 0 MODIFICATIONS 6.0 WALKDOWN OBSERVATIONS 7.0 SURVEILLANCE AND MAINTENANCE 8.0 QUALITY ASSURANCE PROGRAM AND SELF-ASSESSMENT 9.0 EXIT MEETING APPENDIX A LIST OF DEFICIENCIES AND OBSERVATIONS APPENDIX B - LIST OF ATTENDEES
~
~
~
2
3
4
5
6
6
7
8
10
11
. A-1
.
B-1
0 l
e
EXECUTIVE SUMMARY The NRC station blackout (SBO) inspection, led by the Special Inspection Branch of the Office of Nuclear Reactor Regulation (NRR), was conducted from February 28 through March 4, 1994.
The inspection team consisted of staff from NRR and Region IV and a consultant.
The objective of the inspection was to review the licensee's programs, procedures, training, equipment, systems, and supporting documentation for implementing the SBO rule,
CFR 50.63.
Specifically, the team reviewed the coping duration analysis, SBO coping systems, SBO procedures and training, plant modifications and testing, the quality assurance program, maintenance and surveillance, emergency diesel generator and alternate ac source reliability; did a plant walkdown; and witnessed plant operators performing an SBO scenario on the plant simulator.
The team concluded that the implementation of the SBO rule at Palo Verde was consistent with the licensee's submittals and the NRC staff's safety evaluation.
The team conducted this inspection in accordance with Temporary Instruction 2515/120,
" Inspection of Implementation of Station Blackout Rule Multiplant Action Item A-22."
At the time of the inspection, the Unit I SBO program was fully operational.
The alternate ac power sources, while installed, were not yet connected to Units 2 and 3.
However, the licensee stated that the remaining switchgear modifications and testing for Units 2 and 3 will be completed within the stated schedule accepted by the NRC.
The coping duration analysis showing the Palo Verde Generating Station to be a 4-hour coping plant was technically sound.
The calculations and supporting documentation that established the availability of the SBO coping systems were adequate except in these areas:
(1) the Class IE battery sizing calculation used a nonconservative method for determining the starting current for dc motor operated valves and (2) the fault protection relay settings, when powered by the gas turbine generators, may not be adequate.
The quality of the SBO procedures was generally good except that the emergency operating procedure did not include instructions to restore the emergency diesel generators during an SBO event.
The licensee has committed to resolve these issues.
The Palo Verde staff that interacted with the inspection team was well prepared and technically knowledgeable and was able to respond to the team's questions promptly.
The SBO training for the plant staff was good.
The emergency diesel generator reliability program included appropriate elements and activities to ensure that the selected reliability level was maintained in accordance with the reliability program outlined in Regulatory Guide 1.155,
"Station Blackout."
The team found the testing and surveillance of SBO equipment to be adequate.
The team determined that the quality assurance program specified for the SBO equipment met the guidance provided in Regulatory Guide 1.155 Strengths noted by the team included the quality of the design modification and its installation, administrative controls for maintaining the availability and operability of gas turbine generators, and the self-assessment of SBO implementatio.0 INSPECTION BACKGROUND AND OBJECTIVE
A station blackout (SBO) is the complete loss of alternating current (ac)
electric power to the essential and nonessential switchgear buses in a nuclear power plant.
Because many safety systems required for reactor core decay heat removal and containment heat removal depend on ac power, the consequences of an SBO could be severe.
To address this concern, the Commission issued the SBO rule
CFR 50.63,
"Loss of All Alternating Current Power."
As guidance on acceptable methods for meeting the requirements of the SBO rule, the NRC issued, Regulatory Guide (RG) 1.155,
"Station Blackout."
While the regulatory guide was being developed, the Nuclear Management and Resource Council (NUMARC) formulated guidelines and procedures for assessing SBO coping capability and duration.
"Guidelines and Technical Bases for NUMARC Initiatives Addressing SBO at Light Water Reactors,"
documents the NUMARC recommendations.
The RG 1. 155 endorses the NUMARC guidance as acceptable to the NRC staff for meeting the requirements of 10 CFR 50.63.
The licensee evaluated the Palo Verde Nuclear Generating Station using the requirements of the SBO rule and guidance from NUMARC 87-00 and RG 1.155.
The results of this evaluation were submitted to the NRC for review.
The NRC staff issued safety evaluations accepting licensee's methodology for meeting the SBO rule.
The team used these documents for inspection guidance and conducted this inspection in accordance with Temporary Instruction 2515/120,
"Inspection of Implementation of Station Blackout Rule Multiplant Action Item A-22."
In response to the SBO rule, the Palo Verde station installed two gas turbine generators and associated equipment as the alternate ac source, The alternate ac source was designed to be available within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of the onset of an SBO and would power the equipment necessary to cope with an SBO for the remaining 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> of the coping duration.
The licensee performed an analysis to show that Palo Verde can cope without ac power for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.
This was the last of the pilot SBO inspections by the NRC staff.
The objective of this inspection was to verify the adequacy of the programs, procedures, training, equipment, and systems for implementing the SBO rule.
The details of specific areas reviewed and the team's findings and applicable conclusions are described in Sections 2 through 8, The team has characterized its findings as deficiencies or observations.
Deficiencies are the apparent failure of the licensee (1) to comply with a requirement or (2) to satisfy a written commitment or conform to the provisions of applicable codes, standards, guides, or other accepted industry practices that have not been made legally binding requirements.
Observations are items considered appropriate to call to management attention but have no direct regulatory basis.
Deficiencies and observations are listed in Appendix A.
2.0 COPING DURATION ANALYSIS The SBO rule requires systems to have sufficient capacity and capability to ensure that the core is cooled and appropriate containment integrity is
l I
maintained in the event of an SBO for the specified duration.
The calculation of the specified duration was based on factors such as the offsite power design characteristics, emergency alternating current (ac)
power supply configuration, and the emergency diesel generator reliability.
A minimum acceptable SBO duration of 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> was calculated for the Palo Verde, Units 1, 2,
and 3.
The coping duration analysis was neither documented nor initially available for the team's review.
The licensee prepared the documentation during the inspection.
The team verified the following key factors used to determine the SBO duration using the guidance given in NUMARC 87-00 and RG 1.155.
The 4-hour SBO duration was based on an offsite power design characteristic group of "P1,"
an emergency ac configuration group of "C," and an emergency diesel generator (EDG) reliability target of 0.95.
The offsite power design characteristic group "PI" was derived from an independence of offsite power characteristic group of "I2," severe weather group of "1," extremely severe weather group of "2,"
and an expected loss of offsite power of less than
per 20 years.
The team reviewed plant documentation to verify that these characteristics were appropriate.
The team reviewed the EDG reliability data up to February 1992.
The reliability values (unit average)
for the last 100 valid start and load run demands were 1.0 for Units
and 3 and 0.99 for Unit 2.
The reliability data for the last 50 valid start and load run demands were 1.0 for Units
and
and 0.98 for Unit 2 and for the last 20 valid start and load run demands were 1.0 for Units
and 3 and 0.95 for Unit 2.
The team determined that a 0.95 EDG reliability target value was appropriate.
The team concluded that the calculated minimum acceptable SBO duration of 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> was consistent with the NRC's safety evaluation.
3.0 STATION BLACKOUT COPING SYSTEMS 3. 1 Condensate Inventory The team reviewed the licensee's calculation 13-MC-CT-306,
"Condensate Inventory for Decay Heat Removal During a Station Blackout" (Revision 0), to determine if Palo Verde had adequate condensate inventory for decay heat removal during the SBO event.
The calculation was performed in accordance with the guidance in NUMARC 87-00, Section 7.2. 1.
The team noted that, during an SBO event of 4-hour coping duration, the plant would be maintained in hot standby mode in accordance with the SBO Procedure 41EP-IR007,
"Station Blackout" (Revision 1).
The condensate inventory calculation indicated that 84,056 gallons of condensate were required for the decay heat removal during an SBO event of 4-hour coping duration.
The Palo Verde technical specifications required a
minimum useful volume of 300,000 gallons of condensate to be maintained in the condensate storage tank, Thus, the available condensate inventory was greater than the inventory required for the decay heat removal during an SBO event at Palo Verd I
The team concluded that the condensate inventory at Palo Verde was adequate to cope with a 4-hour SBO event.
3.2 Station Battery Capacity The team reviewed the station battery sizing calculation to assure that the battery can power the required SBO equipment for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.
The alternate ac power source will restore power to the battery charger within one hour of the onset of an SBO.
The team verified that the battery sizing calculation (Calculation No.
13-E-PK-202, Revision 6) included appropriate correction factors for sizing the battery.
The team also verified that the input current requirements for the uninterruptible power supply loads were appropriate.
However, the team noted that the licensee used a nonconservative method for determining the starting current for dc motor operated valves (HOVs).
For example, a starting current of 20.68 amperes was used in the calculation for HOV J-AFA-HV-32, whereas the team calculated the starting current to be 45. 18 amperes.
Since the battery was sized for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and a minimum margin of 23.3 percent was available, the team determined that the battery will have adequate capacity to operate SBO equipment for at least 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.
However, the battery capacity margin will be lower than the licensee calculated and will need further evaluation.
The team noted that the licensee's design verification process failed to identify the error in the calculation methodology.'his was identified as Deficiency 94-201-Dl.
The licensee committed to revise the battery sizing calculation using the industry-accepted practice for determining the starting current for HOVs and to revise the relevant documents and surveillance tests to reflect the changes, as required.
The licensee stated that the Class 1E batteries were replaced recently with AT&T round cell batteries and their capacity increases with age.
The team concluded that the battery had adequate capacity to power the SBO loads for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.
A battery charger for one division of equipment would be powered by the alternate ac power source within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of an SBO event.
3.3 Auxiliary Feedwater and Steam Relief To establish reactor decay heat removal during an SBO event, certain auxiliary feedwater and steam relief valves require remote/manual operation.
The team reviewed the heatup calculation, 13-NC-HJ-257, "Ventilation During Station Blackout" (Revision 0), to determine if this equipment would be operable during an SBO event.
This calculation predicted the peak temperatures in the areas where equipment required to cope with an SBO was located.
The team noted that the auxiliary feedwater system flow control valves at Palo Verde were dc-powered HOVs and would be operable during an SBO event.
The atmospheric steam dump valves, which were air operated during normal power operation, were equipped with air accumulators and backup nitrogen gas to facilitate remote/manual operation after a loss of instrument air.
In addition, the atmospheric steam dump valves were equipped with handwheels to allow for manual operation, if needed.
The team verified that the SBO
I J
I
procedures gave adequate instructions for operating these valves.
The staff also found the training given to the plant staff in operating these valves was adequate.
The auxiliary feedwater flow control valves and the atmospheric steam dump valves are located in the top level of the main steam support structure and in the steam turbine driven auxiliary feedwater pump room, respectively.
The calculated peak temperatures were found to be 138'
in the steam turbine driven auxiliary feedwater pump room and 126'
in the top level of the main steam support structure.
These calculated peak area temperatures were found to be less than the temperature limits described in NUMARC 87-00 for equipment operability.
The team concluded that the auxiliary feedwater flow control valves and the atmospheric steam dump valves would be operated remote/manually for decay heat removal.
The team also concluded that procedures and training for operating the auxiliary feedwater and steam relief systems during an SBO event were appropriate.
3.4 Compressed Air The air compressors for the instrument air system would not be available during an SBO event.
The team verified that, except for the atmospheric steam dump valves, no air-operated equipment was required to cope with an SBO event at Palo Verde.
As stated in Section 3.3, the atmospheric steam dump valves were equipped with air accumulators and backup nitrogen gas for operation during an SBO.
The team reviewed the licensee's calculation 13-MS-A20,
"Compressed Gas System Evaluation and Analysis for PVNGS Unit I, 2 and 3" (Revision 3),
and determined that the nitrogen gas supply was adequate for the remote/manual operation of the atmospheric steam dump valves during an SBO event.
In addition, the atmospheric steam dump valves were equipped with handwheels to allow for local/manual operation, if needed.
The team concluded that the equipment for coping with an SBO event would operate after a loss of instrumentation air.
3.5 Effects of Loss of Ventilation During an SBO event, ventilation systems for areas containing equipment required to mitigate the consequences of the SBO event would be lost.
The SBO rule required licensees to identify plant areas that contain equipment required to operate during an SBO and that are susceptible to a significant heatup after an SBO.
The team reviewed (I) calculations identifying the dominant areas of concern, and (2) evaluations showing the reasonable assurance of operability for equipment in these areas.
The team also reviewed calculations showing the structural heat sinks and their associated thermal characteristics and the heat loads for the dominant areas of concern.
The team found that all the peak area temperatures calculated for the SBO event, except the temperatures for the steam-turbine-driven auxiliary feedwater pump room and the top level of the main steam support structure, were less than 120'.
The calculated peak temperatures in the steam-turbine-
t
0 I
l
f
driven auxiliary feedwater pump room and in the top level of the main steam support structure were 138 F and 126', respectively.
These calculated peak area temperatures were acceptable, as discussed in section 3.3.
Therefore, the team found a reasonable assurance of operability for equipment required to cope with an SBO event at Palo Verde.
The team also reviewed calculation 13-HC-HC-303,
"Containment Temperature During Station Blackout" (Revision 0).
The team verified that the temperature profile during an SBO event was bounded by the temperature profile from a
design-basis accident.
The team concluded that the calculations to identify areas of concern were technically sound and that the equipment located in the areas of concern had a
reasonable assurance of operability.
3.6 Containment Isolation Valves The team reviewed both the exclusion process by which the licensee identified the containment isolation valves not required to be controlled during an SBO event and the Updated Final Safety Analysis Report Tables 6.2.4-1'and 6.2.4-2, which listed all the containment isolation valves.
The review indicated that all of the containment isolation valves at Palo Verde conformed with the exclusion criteria described in NUHARC 87-00 and met the intent of RG 1.155.
The. team concluded that adequate containment integrity was assured during an SBO event of 4-hour coping duration at Palo Verde.
3.7 Reactor Coolant Inventory The team reviewed the licensee's calculation showing that reactor coolant system inventory was adequate to cool the core for an SBO event.
Calculation SA-ALL-NCR-88-012, "Station Blackout Coping Analysis for Hargin to Core Uncovery" (Revision I), showed that the Palo Verde units can sustain an SBO event with 120 gpm leakage for 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> without core uncovery.
The calculation was acceptable and conformed with the criteria described in NUHARC 87-00
'alo Verde is a 4-hour coping plant.
During the first hour, the flow paths to and from the reactor coolant inventory would be completely isolated, except paths of the normal leakage.
The emergency operating procedure (EOP) directs the plant operators to connect the alternate ac power source to the blacked-out unit.
The procedure also directs the operators to restore power to the HOVs needed to realign the flow path for one charging pump to supply makeup water to the reactor coolant system.
The team concluded that the procedures, controls, and evaluations of the reactor coolant inventory were adequate to maintain the core cooling for a 4-hour SBO event.
3.8 Emergency Lighting and Communications The team noted that the "Appendix R" emergency lighting system was used to meet the requirements of the SBO rule.
In addition, emergency lights with
I
1-hr-capacity battery packs were installed in the gas turbine generator control room.
Portable lights were also available for the critical SBO equipment areas.
The team verified that the emergency lighting system met the requirements of the SBO rule.
The team verified that adequate communication capability existed for SBO conditions.
Battery backed plant telephones with 8-hour standby capability were the primary means of communication.
Other communication equipment used during an SBO event included a sound powered phone system, a security radio system, a water reclamation facility radio system, and a public address system.
The team noted that the backup system for the regular phones was the sound powered phone system.
This system can only be used within the plant and cannot communicate with the SBO gas turbine generator control room.
Communication with the gas turbine generator control room would still be possible through the security radio system.
The present control room radio, depending entirely on ac power, cannot be used for communications until the gas turbine generators are on line.
A future modification, committed to by the licensee in a letter dated March 20, 1992, will install a new battery-backed radio system in the control room.
The team concluded that the communication equipment would support operations personnel during an SBO.
4.0 PRQCELIURES AND TRA1N1NG 4.1 Severe Weather and Restoration Procedures The team verified that the guidance of Procedure 41A0-1Z158,
"Severe Weather,"
provided adequate instructions to prepare the site for severe weather, including guidance for entry conditions, event control, plant stabilization, and exit conditions.
The instructions for minimizing potential missile hazards and expediting the restoration of important plant systems and components to service were consistent with the guidance in NUMARC 87-00, Section 4.2.3.
The team reviewed the black-start recovery procedure,
"Arizona Public Service Black-Start System Restoration Procedure 1993," dated February 1993.
The procedure instructed Arizona Public Service Company load dispatchers that the highest priority for black-start (offsite source)
be given to restoring power to the Palo Verde Nuclear Generating Stations Other restoration activities such as circuit breaker and relay operations were described in emergency operating procedure.
The team determined that the severe weather and restoration procedures were consistent with the guidance in NUMARC 87-00 except for the deficiency as noted in Section 4'.
4.2 Emergency Operating Procedures The team reviewed emergency operating procedure (EOP)
"Blackout" (Revision 1),
and noted that the EOP did not address the restoration. of the emergency diesel generators (EDGs) during an SBO.
RG 1.155, Section 1.3 and
J
I t
NUMARC 87-00, Section 4.2.2 provide guidance for restoring the emergency ac power system (EDGs) during an SBO.
During the simulation of an SBO, observed by the team, the simulator operating crew did not attempt to restore the EDGs.
This was identified as Deficiency 94-201-D2.
The licensee committed to revise the EOP to state that attempts will be made to restore the EDGs concurrent with restoration of offsite power during an SBO event.
The team also verified that the EOP provided proper instructions for actions such as starting gas turbine generators and energizing the safety bus within
hour of the declaration of an SBO, stripping of ac buses prior to loading gas turbine generators, restoring an offsite ac source, confirming reactor trip and containment isolation, checking auxiliary feedwater initiation and status of shutdown systems, and verifying that proper natural circulation exists for decay heat removal.
4.3 Reliability Program 4.3.1 Emergency Diesel Generator The team reviewed the emergency diesel generator (EDG) reliability program to verify that the EDG reliability data were being trended and the program was consistent with the guidance of RG 1.155.
Palo Vera't,ation Procedures 70PR-ODG01,
"Emergency Diesel Generator Reliability Program" (Revision 1),
"Emergency Diesel Generator Reliability Guidelines" (Revision 1)
and 40DP-90P08,
"Diesel Generator Test Records" established the elements of the reliability program.
The team noted that reliability target level used in the reliability program was consistent with the plant category and coping duration selected.
The EDG surveillance testing was performed in accordance with TS 4.8.1.1.2.
Performance of preventive and corrective maintenance activities was consistent with the operating history and past maintenance activities, vendor recommendations and the results of surveillance testing.
All EDG starts and load runs were monitored and trended by the system engineer.
Failure rates were trended and trigger values established.
Root cause investigations and appropriate corrective actions were initiated for any diesel failures.
Performance indicators, including EDG reliability and availability, were reported to Palo Yerde management.
The team noted that the program adequately identified responsibilities for the major program elements and that a
management oversight program reviewed reliability levels.
The team reviewed the fourth quarter report,
"EDG Reliability Table," to identify exceeded trigger values or problem EDGs and found that all target reliability values
.were met.
The team concluded that the EDG reliability program was detailed and consistent with the guidance of RG 1.155, Section 1.2.
4.3.2 Alternate AC Source The team also reviewed the alternate ac source reliability program to verify that the reliability of the alternate ac power source meets the criteria
i
I I
e j
specified in NUHARC 87-00, Appendix B, Criterion B. 13(a).
The gas turbine system consists oF two redundant gas turbine generators.
As long as the reliability of the gas turbine generator is maintained at or above 0.7764, the system reliability will be maintained at 0.95 or greater.
The following data sets were calculated to give a 95 percent confidence that gas turbine generator reliability is greater than or equal to 77.64 percent:
1 failure in the last 20 valid demands 2 failures in the last 26 valid demands 3 failures in the last 33 valid demands 4 failures in the last 39 valid demands The team noted that the gas turbine generator system had three failures in the last 20, 26, 33, and 39 valid demands'll three failures were due to component failures with gas turbine generator
¹2.
The licensee had taken prompt corrective actions and had successfully retested,the unit several times.
The team noted that the gas turbine generator reliability program was similar to the EDG reliability program.
The licensee has implemented a test program for the gas turbine generators that will have 20 unloaded start tests and
start and load run tests per year.
The team observed the procedure 550P-OGT01,
"Gas Turbine Generator
¹1 Operating Instructions" (Revision 3), directed the operator to adjust the voltage and frequency to approximately 14.2 kV and 60 Hz, whereas the note on page 21 of the procedure stated that the normal values for voltage and frequency to be 12.42 to 14.47 kV and 58.8 to 61.2 Hz, respectively.
The team noted that the SBO loads were analyzed at 13.8 kV.
The licensee revised the applicable procedures to resolve the discrepancy.
The team concluded that the gas turbine generator system reliability program was similar to the EDG reliability program, included trigger values, and was adequate.
4.4 Station Blackout Training The Team reviewed the SBO training given to the operators.
The team noted that the gas turbine generator vendor gave two 5-day training classes for the plant staff.
In addition, the licensee's training department conducted continuing procedure and maintenance training to supplement the vendor training., The team reviewed the training lesson plans and attendance sheets to verify the training process and found the training on SBO procedures and equipment to be adequate.
The team observed the licensee's staff perform an SBO drill on the simulator.
The simulator crew correctly selected the SBO scenario and implemented the SBO EOP 41EP-1R007.
The, simulator crew was able to start the gas turbine generators and load the essential loads in approximately 28 minutes, with the auxiliary operator actions simulated outside the control room.
The licensee's staff performed the EOP on the simulator adequately apart from failing to restore the EDGs, as noted in Section 4.2.
The licensee stated that the
l l
actual performance of the same portion of the EOP at Unit 1 took 42 minutes, well within 'the first hour of coping time.
The procedure was performed by the minimum number of required operators:
one shift supervisor, one control room supervisor, two reactor operators, and four auxiliary operators.
The team reviewed the control room staffing logs during a walkdown of Unit
and verified that each shift was staffed with the required minimum number of operators for the SBO scenario.
The operating staff interviewed were knowledgeable in the SBO implementing procedures and EOPs.
The team concluded that operators were adequately trained and the staffing level was appropriate to cope with an SBO.
5. 0 HOD I F ICATIONS The Palo Verde station committed to install two non-Class 1E gas turbine generators to be used as 1-hour alternate ac power source (only one generator is needed to provide power for an SBO unit).
The team reviewed the significant aspects of the modifications implementing the alternate ac source (modification packages APE-XE-002, APE-XE-003 and 01PE-XE-004).
The team found that the gas turbine generator had a loading margin of 17 percent, corresponding to a maximum loading of 2809 kW.
The new cable installed between the gas turbine generators and the plant was properly sized with regard to short circuit, voltage drop, and ampacity.
A minimum of two breakers allowed the plant to be isolated electrically, protecting the Class IE source, and adequate electrical separation was maintained between the power sources.
The voltage drop study, 13-EC-NE-200 (Revision 2),
was satisfactorily performed.
This study showed that the minimum voltage at the terminals of the largest motor to be started up by the preloaded gas turbine generator was 83 percent of the motor nominal terminal voltage, or 3 percent higher than the minimum voltage required.
The short circuit study performed indicated that the equipment was properly sized.
The gas turbine generator controls and surveillance consisted of microprocessor-based equipment with online diagnostics.
The team considered this equipment an asset.
The microprocessor controls were backed up by electromechanical controls.
A remote alarm was provided in the Unit 1 control room For gas turbine generator or battery trouble and for heating, ventilation, and air-conditioning system trouble.
The lineup of the gas turbine generator source to a unit under SBO would require that the gas turbine generator operations be controlled from the gas turbine generator control room and that breaker closing (13.8 kV breakers at the power block) to connect the source to the plant be performed by the plant operators.
The alternate ac source will provide power to the blacked-out unit, and the 525 kV switchyard loads through Unit 1 non-Class 1E switchgear.
The team noted that the calculation for setting the protective relays (Calculation AO-EC-NA-422 (Revision 0)), did not take into account several important factors, such as the decrement of the short circuit current contribution by the gas turbine generator and the impedance at the point of fault.
Furthermore, relay response depends only on the particular current
I
flowing at the relay location, not on the total fault current.
Failure to consider these factors led to the nonconservative conclusion that the relays would "see" sufficient current for proper operation.
Contrary to this conclusion, the team estimated that, with the existing relay settings, some of the relays might fail to trip and isolate faults.
Uncleared faults could damage equipment or injure personnel.
The licensee has committed to review the relay settings 'and revise the corresponding analysis of relay operations to include the team's findings.
This was identified as Observation 94-201-01.
In implementing the modification packages, the licensee used controlled drawings, documents, and procedures.
Management oversight and project administration for the implementation were good.
The procurement specifications for SBO coping equipment conform with the guidance in RG 1.155.
The
CFR 50.59 evaluations were generally good and showed that this modification did not involve an unreviewed safety question.
The post-modification testing confirmed that gas turbine generators would start and pick up the SBO loads and that the motors would start.
The team also noted that other functional tests performed were consistent with the recommendations in the NRC safety evaluations.
The team performed a walkdown of the installation of modifications for Unit 1, which were complete and operational'he team also inspected the installation being done for Unit 3.
The team found that the installation conformed with the requirements of the specifications and that the intent of the modifications had been met.
In addition, the team reviewed the effects of flash flood on the availability of the alternate ac source.
The review indicated that the gas turbine generators were sufficiently elevated and that the operation of the gas turbine generators would not be affected by the flash floods.
The team concluded that the design package contained acceptable design data, incorporated the guidance provided in NUMARC-87-00, Appendix B, and was technically sound.
The quality of design modification and installation was considered a strength.
6.0 WALKDOWN OBSERVATIONS The team performed a walkdown with an auxiliary operator to simulate the operator actions, as required in EOP 41EP-1R007.
The auxiliary operator appeared knowledgeable in the requirements of SBO and the locations of equipment.
The auxiliary operator properly simulated the procedure steps.
The team also verified the inventory and operability of the portable lights stored in the emergency equipment cabinet at the 140-foot elevation of the control building.
The requirements for portable lighting were provided in the licensee's Procedure 14FT-9FP06,
"Fire Equipment Locker and Emergency Equipment Cabinet Inspection" (Revision 4).
The team found that portable lights were in operable condition and that the material condition of the SBO equipment was good.
l
I l
7.0 SURVEILLANCE AND MAINTENANCE The team verified that the required monthly preventive maintenance procedures for the gas turbine generators and the associated equipment were approved.
The monthly preventive maintenance procedures reflected the vendor maintenance guidelines.
However, the semi-annual and annual preventive maintenance procedures were in draft form.
The licensee stated that these procedures would be approved and issued shortly.
The team noted that the licensee had the approved procedures for the gas turbine generator protective relay testing and calibration.
The team found the protective relay test procedures to be adequate and the tests and calibration performed on these relays were satisfactory.
The team also verified that diesel fuel inventory at the Palo Verde site was sufficient to operate both gas turbine generators for 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.
The team reviewed the daily inventory log sheets of the fuel oil storage tanks, the minimum storage volume requirements, and the gas turbine generator fuel consumptions.
The team concluded that sufficient diesel fuel was available on site to sustain the SBO coping duration.
The licensee has appropriately prioritized maintenance work activities for the gas turbine generators.
Specifically, if a gas turbine generator was rendered nonfunctional, the repair would be considered a Priority 2.
According to the licensee's Procedure 01PR-OAP02,
"PVNGS Priority System," Priority 2 work is defined as actions required to permit or sustain power production, meet external commitments with hard due dates, or return plant systems to compliance with license basis documents.
Procedure 30DP-9WPOI, requires that troubleshooting and/or repair work begin within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of problem identification.
The licensee stated, that if both gas turbine generators are out of service, additional resources would be allocated to return at least one gas turbine generator to service as soon as possible.
Palo Verde's administrative control to maintain the availability and operability of gas turbine generators is consistent with TS 3.8. 1. 1.
Specifically, the licensee established a 24-hour common mode time limit so that, if one gas turbine generator failed to start and the second had not been started within the previous 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, the second gas turbine generator must be started within the next 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to verify that a
common mode failure does not exist.
The team considered Palo Verde's administrative controls for maintaining the availability and operability of gas turbine generators a strength.
8.0 EQUALITY ASSURANCE PROGRAM AND SELF-ASSESSMENT The team reviewed the quality assurance (gA) program for the SBO equipment to verify that the program met the requirements outlined in RG 1. 155, Appendix A.
Appendix F-6 of the Operations guality Assurance Plan implemented the quality assurance program for SBO coping equipment.
The team reviewed Appendix F-6 and found it in agreement with the guidelines of RG 1. 155, Appendix A.
l t
I
The team noted that the gas turbine generator system, including local 13.8 kV switchgear, batteries, control panel, and auxiliaries, was purchased in accordance with Palo Verde Quality Assurance
"QAG" requirements from Solar Turbines as a TurnKey project.
The Solar Turbine Quality Assurance specification met the intent of RG 1. 155, Appendix A.
The cables and 13.8 kV switchgear were purchased with appropriate QA requirements.
The performance testing of gas turbine generators and turbine control room equipment was done at the factory and witnessed by the Palo Verde staff.
A final acceptance test was also performed by the vendor after the installation.
The team observed that according to the procurement document for the gas turbine generators, the generators are rated at 13.8 kV, 3 phase,
Hz at a
minimum rated electrical power output of 3400 kW under the worst environmental conditions of -10'
to 50 CD The actual generator rating from the nameplate was found to be 5338 kVA and the ambient temperature was 40'.
The team questioned the generator rating at 50'.
The kW rating of the generator had not been formally evaluated.
In response, the team was given the derating factor needed for 50'
ambient temperature and supporting documentation that showed a derated generator rating of 3907 kW.
The team found this rating acceptable.
The team reviewed the recently completed self-assessment by the licensee's Independent Safety and Quality Engineering group.
The assessment was thorough and covered all areas of the SBO rule implementation.
Several recommendations were made, and enhancements of procedures and calculations were suggested.
At the time of the inspection, the licensee's findings had been either resolved or were in the plant tracking system awaiting resolution.
The team determined that the licensee management was proactive in identifying and resolving issues.
The team concluded that the QA program for the SBO coping equipment was adequate.
The team considered self-assessment a strength.
9.0 EXIT MEETING The team held an exit meeting with the licensee representatives (listed in Appendix A) at the conclusion of the inspection on March 4, 1994.
During this exit meeting, the team summarized the scope and findings of the inspection.
I
APPENDIX A LIST OF DEFICIENCIES Number 94-201-D1 94-201-D2 Title Inadequate battery calculation Re ort Location 3.2 Inadequate Station blackout procedure 4.2 LIST OF OBSERVATIONS Number 94-201-01 Title Inadequate fault protection Re ort Location I
APPENDIX B
LIST OF ATTENDEES Arizona Public Service Com an M. Eskinazi, Supervisor, Nuclear Engineering A. Fernandez, Mechanical Engineer, Plant Engineering R. Field, Lead System Engineer J.
Hesser, Director, Nuclear Engineering J.
Holmes, Sr'ngineer, Nuclear Engineering D. Kissinger, Supervisor, Independent Safety 8 equality Engineering J.
Levine, Vice President, Nuclear Production S. Hoyers, Supervisor, Maintenance Standards G. Overbeck, Assistant to Vice President, Nuclear Productions RE Pontes, Project Manager, Nuclear Projects R. Prabhakar, Manager, Independent Safety
& equality Engineering R. Rogalski, Engineer, Licensing C. Rogers, Technical Assistant, Licensing J.
Sears, Principal Engineer, Technical equality Engineering T.
Shaw, Manager, WRF usr,nc E.
Imbro, Branch Chief, Special Inspection Branch P. Koltay, Section Chief, Special Inspection Branch K. Johnston, Sr.
Resident Inspector W. Ang, Engineering Branch Chief, RIV R. Hathew, Team Leader F.
Gee, Team member, RIV A. Pal, Team Member, NRR DE Shum, Team Member, NRR 0. Hazzoni, Team Member, Consultant
~ ~
1 \\
'