IR 05000244/1996007

Insp Rept 50-244/96-07 on 960721-0824.Noncited Violations Noted.Major Areas Inspected:Operations,Maint,Engineering & Plant Support

ML17264A705
Person / Time
Site:	Ginna
Issue date:	10/18/1996
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML17264A704	List: IR 05000244/1996007 ML17264A703
References
50-244-96-07, 50-244-96-7, NUDOCS 9610290082
Download: ML17264A705 (64)

Text

'

U. S. NUCLEAR REGULATORY COMMISSION

REGION I

Docket No:

License No:

50-244 DPR-18 Report No:

Licensee:

50-244/96-07 Rochester Gas and Electric Corporation (RGSE)

Facility:

Location:

R. E, Ginna Nuclear Power Plant 1503 Lake Road Ontario, New York 14519 Dates:

July 21 - August 24, 1996 Inspectors:

P. D. Drysdale, Senior Resident Inspector, Ginna C. R. Thomas, Acting Resident Inspector, Ginna C. C. Osterholtz, Operator Examiner, Rill G. S. Vissing, Project Manager, NRR Approved by:

L. T. Doerflein, 'Chief, Reactor Projects Branch

Division of Reactor Projects 96i0290082 9610i8 PDR ADOCK 05000244

PDR

'

EXECUTIVE SUIVIIVIARY R. E. GInna Nuclear Power Plant Ins ection Re ort No. 50-244/96-07 This integrated inspection report includes aspects of licensee operations, engineering, maintenance, and plant support.

The report covers a five week period of resident and project manager inspections.

~Oerations:

On July 22, 1996, the licensee discovered a leak in the containment spray system outside containment.

The leak rate was small, but could not be isolated without declaring all emergency core cooling system (ECCS) trains inoperable.

The plant was operated for approximately 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> while the leakage was evaluated.

The following day, the technical specifications (TS) limiting condition for operation (LCO) 3.0.3 was entered pursuant to an administrative procedure requirement after the leakage was measured to be above the analyzed limitof 2 gallons per hour, and could not be isolated.

However, the leak was isolated using a freeze seal before entry into Mode 3 was necessary.

An unresolved item was opened because the plant was apparently operated outside its analyzed limit prior to the leakage being isolated (URI 50-244/96-07-01).

On August 2, 1996, the unit was shut down to upgrade the residual heat removal (RHR)

core deluge valve motor actuators.

The actuators may not have been capable of overcoming a postulated pressure'ocking condition, and the valves could not be modified with the plant at power.

Just prior to the shutdown, operator error was attributed to the failure of a service water pump to start.

During the. shutdown, weak work and equipment configuration controls resulted in an engineered safety feature actuation when the main turbine was latched for a control valve test.

During the subsequent plant startup, the safety injection (SI) system accumulator test line relief valve RV-887 lifted, and did not reseat properly.

This resulted in Sl accumulator leakage through leaking system valves and RV-887. Although the leakage was initially small, it grew progressively larger.

All leakage was collected in the pressurizer relief tank.

The licensee made various attempts to identify and isolate multiple Sl system leak paths.

The leakage required frequent operation of Sl pumps to maintain the required water level in the Sl accumulators.

The licensee continued to monitor, analyze, and troubleshoot the leakage for approximately three weeks, when the inspection period ended, without resolution.

However, the licensee anticipated a planned shut down to perform repairs.

A temporary modification was also planned to install a dedicated pump to relieve the frequent Sl pump starts.

On August 13, 1996, an existing minor steam leak in the B-main feedwater (MFW)

regulating valve inlet isolation valve increased significantly. After the leak increased, it appeared to represent a potential operational and personnel safety concern.

The licensee attempted to make an on-line seal injection repair of the valve, but the first attempt was not successful.

Engineering developed a plan to perform a more extensive on-line repair;

J Executive Summary (cont'd)

however, this plan was not implemented.'he licensee planned to disassemble and repair the valve during a plant shutdown.

On the afternoon of August 20, 1996, a faulted control signal caused the 8-main feedwater (8-MFW) regulating valve to go full closed, and the reactor automatically tripped when the 8-steam generator level reached the low trip set point.

Control room operators responded to the event effectively.

EOPs were correctly implemented, and the transition to normal operating procedures was performed efficiently. The trip was attributed to a maintenance preventable functional failure in a pressure transducer.

During the shutdown, auxiliary feedwater (AFW) throttle valves did not function as required, An Unresolved item (URI 50-244/96-07-02) was opened to evaluate the minimum AFW flow requirements for power operations, and for accident conditions.

After repairs were made to the 8-MFW pressure transducer, the reactor was returned to power operations within 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />.

No attempts were made to repair the Sl system or feedwater valve leakage.

The NRC considered that returning the plant to power in a short time period with several degraded equipment conditions lacked the conservative philosophy usually noted.

IVlaintenance:

Good communications were noted between maintenance, engineering, and management during the outage to upgrade RHR valve actuators.

Quick and timely resolution of field problems were observed during the equipment upgrades.

The electrical maintenance staff demonstrated proficiency and expertise during installation of the new valve motors, cabling, and breakers.

The licensee did not take the opportunity to disassemble V-3984 after the reactor tripped on August 20, 1996.

However, the maintenance activities for the subsequent on-line leak repair were adequately controlled.

The licensee appropriately avoided going beyond the initial repair so that additional evaluations could be made for further repairs.

Installation and testing of the A-service water (SW) pump motors was effectively performed and was well coordinated.

Maintenance technicians maintained a questioning attitude; and the technical involvement of engineering and management led to the timely resolution of defective conditions internal to the motors.

Good coordination and judgement between maintenance, engineering, and management were noted during this work.

The scope and performance of diagnostic testing on both the A & 8 emergency diesel generators (EDGs) was in accordance with the recommendations from its owner's group consultant.

The licensee is taking a proactive approach in assisting the owner's group develop new guidance that will be included in an upgraded maintenance program for ALCO EDGs.

Results and Test personnel exhibited good awareness in identifying a power anomaly during a shutdown of the 8-EDG after a surveillance test.

Technicians demonstrated good skills in troubleshooting and diagnosing the problem.

The problem was properly diagnosed and appropriate corrective actions were implemente i

Executive Summary (cont'd)

Installation and testing of digital volt meters on instrument buses and Twinco panels was well coordinated and performed in a professional manner, and in accordance with applicable codes and standards.

The initiative to develop a new diesel test methodology using thermography appeared to provide an important improvement in the reliability centered maintenance program for the diesel generators, could improve component reliability through early detection of equipment degradation, and could validate preventive maintenance frequencies.

Surveillance testing activities were thorough and well coordinated, personnel properly adhered to procedures, and technicians exhibited good technical knowledge.

A main feed pump (MFP) breaker was not properly configured in the test position during a plant startup which resulted in an automatic start of the A auxiliary feedwater (AFW)

pump, an engineered safety feature actuation, when the AFW bypass switches were placed in "Normal." The cause of the improperly configured MFP breaker was personnel error.

~En ineerin The design analyses and safety review for seismic and pressure boundary qualification of the freeze seal used to repair the containment spray leakage employed conservative engineering assumptions, and provided adequate bases to show that the freeze seal installation did not place the system outside any design basis limitations.

The initial on-line repair to seal a MFW isolation valve leak was partially successful.

However, the licensee appropriately determined that a second repair method that involved drilling holes in the valve body may not be effective, and that corrective action should be performed during an outage when the valve could be dissembled and repaired.

The licensee has established a good 10 CFR 50.59 determination process.

It is highly dependent on the training of people preparing and reviewing the Safety Reviews and preparing and reviewing Safety Evaluations.

The training was assessed as very good and was accomplished by a structured training program in which many engineers have been trained.

Many engineers have been certified as qualified reviewers.

Only minor program discrepancies were noted.

The licensee's evaluation of the radiation monitor near the spent fuel pool to detect an inadvertent criticality in the new fuel preparation area is ongoing.

The licensee will either verify the existing monitor meets the requirements or request an exemption from 10 CFR 70.24 prior to the next time new fuel is stored onsite.

The licensee took prompt action to clean areas in the auxiliary building that could potentially cause inadequate ventilation of electrical equipment.

The procedure for

Executive Summary (cont'd)

inspecting the material condition of electrical equipment was adequate; however, a review of its frequency for safeguards buses was appropriately undertake '

TABLE OF CONTENTS EXECUTIVE SUMMARY.. ~... ~.....

~

~

~

~

~

~

~

~

~

II TABLE OF CONTENTS v

I. Operations

~

~

~

~

~

~

~

01 Conduct of Operations..............................

~ ~......

01.1 General Comments

01.2 Summary of Plant Status....... ~... ~..... ~... ~...

~. ~......

02 Operational Status of Facilities and Equipment..... ~...... ~.....

~

~.

02.1 Power Reduction Due to Containment Spray System Leak...........

02.2 Plant Shutdown Required to Upgrade Motor Actuators For Residual Heat Removal (RHR) Core Deluge Valves MOV-452A and MOV-452B....

~.

02.3 Safety Injection System Accumulator Leakage.............

~..,..

02.4 Feedwater Block Valve V-3984 Leakage..........

02.5 Automatic Reactor-Trip Due to Low Steam Generator Level Following Isolation of the B-Feedwater Train...........................

04 Operator Knowledge and Performance...............

04.1 Failure of The A-SW Pump to Start

. ~..........

~. ~..........

~

04.2 AFW Pump Automatic Start After Latching the Main Turbine for Control Valve Testing.............................

~....

08 Miscellaneous Operations Issues....'....... ~.......... ~....

08.1 (Closed) LER 96-008: Main Feedwater Pump Breakers Open, Due to Low Seal Water Differential Pressure, Results in Automatic Start of the A-AuxiliaryFeedwater (A-AFW) Pump.......................

08.2 (Open) LER 96-009: Leak Outside Containment, Due to Weld Defect, Results in Leak Rate Greater Than Program Limit

I. Maintenance

.

~

~

~. ~..............................

~.. ~.........

I M1 Conduct of Maintenance.....

M1.1 Maintenance Observations

. ~.....

~.. ~........

~..

~

~ ~.......

M1.1.1 Installation of New Motors and Cables for MOV-852A and MOV 85 OV 852B

'..

M1.1.2 On-line Leak Repair of Main Feedwater Valve V-3984........

M1.1,3 Installation Testing of New Service Water (SW) Pump Vacuum Breaker and Motor.......... ~..............,

. ~.....

M1 ~ 1.4 Emergency Diesel Generator (EDG) Diagnostic Testing.......

M1.1.5 Emergency Diesel Generator (EDG) Testing and Troubleshooting

M1.1.6 Installation of Meters on the Instrument Buses And Twinco Panels 24 M1.1.7 A-Emergency Diesel Generator (EDG) Thermography Survey...

M1.2 Surveillance Observations

M4 Maintenance Staff Knowledge and Performance.............. ~....

M4.1 Autostart of A-AFW Pump Due to MFP Breaker Not in The Test Position

VI

Table of Contents (cont'd)

III. Engineering

E2 Engineering Support of Facilities and Equipment..

~ ~.....

~

~..

~ ~....

E2.1 Design Analyses and Safety Review for Seismic and Pressure Boundary Qualification of Freeze Seal Equipment Used to Isolate Containment Spray Leakage

E2.2 Analysis of Main Feedwater Valve V-3984 Temporary Leak Repair....

E7 Quality Assurance in Engineering Activities ~....,,

. ~...

~

~. ~...

~

~

~

~

E7.1 Review of the Licensee's 10 CFR 50.59 Program.......

IV. Plant Support R1 Radiological Protection and Chemistry (RP5C) Controls R1.1 Radiation Monitors Near the New Fuel Storage Area F1 Control of Fire Protection Activities F1

~ 1 Housekeeping in the Auxiliary Building

~

~

~

~

~

35

36

~

~

~

~

~

V. Management Meetings X1 Exit Meeting Summary X3 Management Meeting Summary........ ~....

X3.1 Senior NRC Management Visits to The Ginna Station

~

~

~

~

~

~

~

~

~

~

37

~

~

~

~

~

Re ort Details I. ~Oeratiane

Conduct of Operations'1.1 General Comments (Inspection Procedure (IP) 71707)

The inspectors observed plant operations to verify that the facility was operated safely and in accordance with licensee procedures and regulatory requirements.

These reviews included tours of the accessible areas of the facility, verification of engineered safeguards features (ESF) system operability, verification of proper control room and shift staffing, verification that the plant was operated in conformance with technical specifications and that appropriate action statements for out-of-service equipment were implemented, and verification that logs and records were accurate and identified equipment status or deficiencies.

01.2 Summar of Plant Status a.

Ins ection Sco e (71707)

At the beginning of the inspection period on July 21, 1996, the plant was operating at full power.

On July 22, 1996, the licensee discovered a leak in the containment spray system outside containment.

The leak rate was small, but could not be isolated from the refueling water storage tank (RWST). The licensee continued to operate the while the leakage was evaluated.

The following day, the licensee entered technical specifications (TS) limiting condition for operation (LCO) 3.0.3 pursuant to an administrative procedure requirement after determining that the leakage exceeded the analyzed limit of 2 gallons per hour (gph) ~ A power reduction was initiated and the licensee proceeded toward cold shutdown to repair the leak.

However, the licensee was able to isolate the leak using a freeze seal on the piping before entry into Mode 3 was required.

After the leak was isolated, the plant was returned to power, and the defective pipe section was removed and replaced.

During an NRC inspection (50-244/96-08) from July 22 to August 8, 1996, of RGSE's Generic Letter (GL) 89-10 program, the licensee was not able to confirm that the motor actuators on the residual heat removal (RHR) core deluge valves were capable of overcoming potential pressure locking conditions.

Consequently, on August 2, 1.996, the unit was shut down to upgrade the actuators.

During the subsequent plant startup, the safety injection (Sl) system accumulator discharge header relief valve RV-887 lifted when Sl system check valve

'opical headings such as 01, M8, etc., are used in accordance with the NRC standardized reactor inspection report outline.

Individual reports are not expected to address all outline topics.

The NRC inspection manual procedure or temporary instruction that was used as'nspection guidance is listed for each applicable report sectio backleakage from the reactor coolant system (RCS) caused Sl system pressure to exceed the relief valve set point.

The licensee was able to properly re-seat the Sl check valves, but RV-887 did not reseat and continued to leak at a low rate.

Although the leakage was initially small, it grew progressively larger and eventually reached approximately 0.75 gpm by the end of the inspection period.

The licensee made'various attempts to identify and isolate multiple leak paths; however, RV-887 continued to leak throughout the inspection period.

The leakage required frequent operation of Sl pumps to maintain the required water level in the Sl accumulators.

At end of reporting period, the licensee planned to identify all Sl valves that were leaking, and was anticipating a planned shut down to perform repairs.

The licensee also planned a temporary modification to install a separate dedicated pump to relieve the frequent Sl pump starts.

On August 13, 1996, an existing minor steam leak on the B-main feedwater regulating valve (FRV) inlet isolation valve increased significantly. After the leak increased, it appeared to represent a potential operational and/or personnel safety concern.

The licensee attempted to make an on-line seal injection repair of the valve, but the first attempt at this repair was not successful.

Engineering developed a plan to perform a more extensive on-line repair; however, this plan was not implemented.

At the end of the inspection period, the licensee planned to disassemble and repair the valve during the anticipated plant shutdown noted above.

On the afternoon of August 20, 1996, with the plant at 100% power, the B-FRV control system received a faulted input signal that caused the valve to go full closed.

Operators attempted to restore full feedwater flow to the B steam generator (SG), but were not able to recover flow before SG level decreased to the low level trip point (17%) and the reactor automatically tripped.

The cause of the trip was a loss of electrical continuity in a pressure transducer that provides an input signal to the automatic digital feedwater control system (ADFCS). After repairs were made to the transducer, the reactor was returned to power operations the following day and remained at full power until the end of the inspection period on August 24, 1996.

Operational Status of Facilities and Equipment Power Reduction Due to Containment S ra S stem Leak Ins ection Sco e (71707)

On July 22, 1996, the licensee discovered a small leak in a containment spray system pipe elbow outside containment.

The inspectors reviewed the licensee's evaluation of and response to the leakage, and the subsequent pipe repairs.

Observations and Findin s At approximately 11:30 a.m. on July 22, 1996, a radiological protection technician working inside a locked high radiation area in the auxiliary building discovered a leak

in a containment spray (CS) system pipe.

The leak appeared to be from a weld on an elbow joint in a 2 inch test line from the RWST to the sodium hydroxide (NaOH)

chemical eductors.

The test line is only used during periodic surveillance testing to avoid injecting NaOH into the main CS flow stream that recirculates back to the RWST. The leak was determined to be a through wall, pinhole size flaw in the weld joint and was emitting a fine spray of borated water under an RWST head pressure of approximately 40 psig.

The leak could not be isolated without declaring the RWST inoperable and shutting its discharge valves.

An inoperable RWST would require entry into TS LCO 3.0.3 and shutting down the reactor.

The licensee took actions to contain the leakage within a polyethylene wrap around the elbow joint, so that the leakage could be directed to a bucket where a rough estimate of the leak rate could be made.

The water was then diverted to a plant collection system.

Initially, the licensee did not attempt to precisely quantify the leakage, but a rough estimate of 2 - 3 gph was obtained using the collection bucket.

The reactor remained operating in Mode 1 for at least 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> while engineering analyses and operability reviews were performed on the RWST water supply to the emergency core cooling systems (ECCSs)

~ Engineering analyzed the consequences of a complete failure of the test line on ECCS injection capability.

These evaluations concluded that a double ended break of the test line would still permit sufficient ECCS injection volume from the RWST to supply the minimum coolant inventory required for design basis accident conditions, and would also provide a sufficient water level in the containment sump to permit transition to recirculation cooling.

The licensee concluded that the existing leakage did not render the RWST or any ECCS equipment inoperable.

Engineering also performed a structural integrity evaluation and determined that the seismic qualification of the defective joint was maintained.

The use of a freeze seal on the test line as a means of isolating the leak was also evaluated so that a repair could be made without taking the RWST out of service.

Operations personnel and the plant operations review committee (PORC) noted that water level and boron.

content were the only specific criteria in technical specifications for RWST operability.

However, shutting the two main RWST discharge valves would have isolated the leak, but it would have rendered all ECCS flow trains inoperable, and would have required an immediate entry into LCO 3.0.3 in accordance with the plant TS.

For the remainder of the day, the licensee continued to evaluate various repair options and the potential consequences of the leak.

The following morning, Results and Test Department personnel recognized that an analyzed limit of 2 gph existed for integrated ECCS system leakage outside containment.

These requirements. were contained in administrative procedure A-52.4, "Control of Limiting Conditions for Operating Equipment," and test procedure PT-39, "Leakage Evaluation of Primary Coolant Sources Outside Containment."

Consequently, at 8:00 a.m., the licensee initiated PT-39 to perform an accurate measurement of the leak rate.

Until that time, operations personnel and the PORC were not cognizant of the limit because it was not explicitly continued in the TS, but was placed in administrative procedure A-52.4 during the conversion to

improved technical specifications (ITS) in February 1996.

Also, PT-39 was normally performed only once per year, and its requirements were also not readily known by most plant personnel.

Procedure ECA-1.2, "LOCA Outside Containment," is the only other plant procedure that requires operators to take actions to deal with ECCS leakage outside containment.

However, it assumes that a loss of coolant accident (LOCA) outside containment has already occurred and is causing a loss of RCS inventory.

No other normal or abnormal operating procedure directs any actions based on ECCS leakage outside containment.

After performing PT-39, the licensee confirmed that the leakage was above 2 gph (total CS leakage was 2.599 gph and total integrated ECCS leakage was 2.742 gph) ~

Operations immediately entered LCO 3.0.3 and initiated a plant power reduction to be in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, Operators effectively controlled the plant during the power reduction.

During the downpower, engineering completed a

safety analysis for use of a freeze seal, and a freeze seal was subsequently installed.

Shortly thereafter, the leak was isolated and entry into Mode 3 was not necessary, Consequently, the licensee removed and replaced the leaking elbow and adjacent piping while returning the plant to full power.

The inspectors noted that Chapter 15 of the Ginna UFSAR (Section 15.6.4.3) and procedure A-52.4 both contained the 2 gph limit due to design and licensing bases criteria.

Leakage less than or equal to 2 gph will maintain exposures at the site boundary and in the low population zone below the maximum 10 CFR 100 limits during the recirculation cooling phase following a design basis accident where maximum fuel damage is postulated.

A licensee event report (LER 96-009) was issued on August 22, 1996, and indicated that the CS system leakage was outside the analyzed limit. The licensee also indicated that the preliminary results of a Westinghouse analysis in progress show that leakage from the CS test line could be limited to 30 gph because it would be isolated during a large break LOCA, and that a 30 gph leak would not produce exposures in excess of 10 CFR 100 limits. The licensee also indicated that the ITS basis for LCO 3.6.1, "Containment," would be revised so that leakage of this type would involve a timed action statement for containment operability, so that entry into LCO 3.0.3 would not be required.

The inspectors noted that the actual total ECCS leakage of 2.742 gph was outside the limit stated in the UFSAR. Subsequent analysis is expected to show that this leakage would have kept offsite exposure below 10 CFR 100 limits. This item will remain unresolved pending NRC review of additional design basis accident analyses, and revisions of the UFSAR and ITS (URI 50-244/96-07-01).

Conclusions The inspectors concluded that'the licensee should have completed a detailed leak rate measurement before 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> had elapsed with the plant at full power.

Placing the leakage limit and LCO entry condition in an administrative procedure, that were not also explicit in the ITS, caused a delay in actions to obtain an accurate

measurement of the leak flow rate, to bypass a requirement to enter LCO 3.0.3.

Although entry into LCO 3.0.3 may have been a conservative program requirement, the program limit of 2 gph was derived from a USFAR accident analysis.

Subsequent analysis by Westinghouse is expected to provide a basis for updating the accident analysis, and show that the specific leakage in this case would not have generated offsite exposure levels above 10 CFR 100 limits. This item remains unresolved pending further review of the licensee's analysis, and future UFSAR and ITS changes.

02.2 Plant Shutdown Re uired to U rade Motor Actuators For Residual Heat Removal RHR Core Delu e Valves MOV-452A and MOV-452B at Ins ection Sco e (71707)

During an NRC inspection from July 29 through August 2, 1996, of RG&E's Generic Letter (GL) 89-10 Program, the licensee was not able to demonstrate that adequate

'thrust margin existed to assure that the residual heat removal (RHR) core deluge

=

valves MOV-852A and MOV-852B could open to perform their safety function under potential pressure locking conditions.

(refer to NRC GL 95-07 and inspection report 50-244/96-08).

The inspectors reviewed operational activities related to this shutdown.

b.

Observations and Findin s At 6:45 p.m. on August 2, 1996, operators initiated a power reduction and proceeded toward cold shutdown so that upgrades could be made to the core deluge valve actuators.

The plant shutdown was necessary to upgrade the motor actuators on valves MOV-852A 6 B, which are part of the low pressure ECCS injection flow path to the reactor vessel.

These flow trains could not be removed from service, at power since the valves are required to be operable in Modes 1, 2, and 3. The plant entered Mode 2 at 12:42 a.m. on August 3, 1996, and Mode 3 at 12:46 a.m..

At approximately 1:00 a.m., after tripping the main turbine, steam dump valve AOV-3355 did not open as expected.

This was reportedly due to a sticking pilot valve.

RCS T-avg subsequently increased and reached 553 degrees Fahrenheit (

F)

when the A-atmospheric relief valve (A-ARV) lifted on high RCS pressure as designed.

When RCS pressure dropped below the ARV setpoint, the ARV closed normally.

During the outage, the steam dump pilot valve was disassembled and cleaned, but no cause for its faulty operation was found.

Also at 1:00 a.m. on August 3, 1996, when the auxiliary feedwater pumps were started for SG cooling, both AFW discharge throttle valves MOV-4007 and MOV-4008 did not throttle properly to limit AFW flow to less than 230 gpm as required when the pumps were started.

The AFW flow must throttle to less than 230 gpm to prevent overfeeding a SG in the event of a high energy line break (see section 02.5 below). At 7:56 a.m. on August 3, 1996, the plant entered Mode 4 when T-avg reached less than 350 During the outage on August 6, 1996, with the plant in Mode 4, ISC technicians made adjustments and conducted testing on the main turbine ¹2 control valve in accordance with maintenance procedure M-109, "AEH Governor High Pressure Fluid System Adjustment, Calibration, and Maintenance."

The auto-start function of the AFW system was enabled on the main control board (MCB) because operators anticipated escalating power soon through Modes 2 and 1 where the autostart function is required to be available by the ITS. When operators "latched" the main turbine, both AFW pumps started (an ESF Actuation) ~

The licensee subsequently made a four-hour ENS call at approximately 4:00 p.m. per 10 CFR 50.72 due to the ESF actuation.

Operators were able to control AFW flow to both SGs following the event without a significant impact on the plant (see section 04.2 below).

The IS.C work package had been approved by the operations staff on shift who understood that the work required latching the main turbine from the main control board.

However, the operators did not fully understand the scope or effect of the test on plant systems, since they did not anticipate that latching the main turbine with both MFP breakers. open would cause an AFW auto-start.

After this event, the operations manager directed that all ongoing outage work should be stopped, and processed through the operations office before being resumed.

In an effort to obtain reliable automatic throttling of MOV-4007 and MOV-4008, the licensee approved a temporary plant modification to throttle the AFW pump discharge manual valves V-4011 and V-4012 to 240 gpm each.

The licensee throttled V-4011 & V-4012 and then tested the automatic throttle capability of MOV-4007 and MOV-4008. After the plant was returned to hot shutdown on August 7th, the throttling appeared to work well because test results were satisfactory and showed good repeatability (within a 2 gpm).

During the RCS heatup at 4:00 a.m. on August 7, 1996, and with the reactor in Mode 3, the B-reactor coolant pump (B-RCP) failed to start when directed by operators from the MCB. IVlaintenance electricians inspected the pump breaker and a second start was attempted, but was also unsuccessful.

A white light ("breaker not in required position") on the MCB switch was received on both attempts to start the pump.

Further troubleshooting found a faulty overcurrent relay in the breaker.

After replacing the relay, the breaker worked correctly and a third attempt was successfully made at approximately 7:00 a.m..

During the plant heat up on August 7, another autostart of the A-AFW pump occurred at 8:11 p.m. when in Mode 3 because a MFP breaker was not in the proper test position (see para M4.1 below),

During the initial approach to criticality later that evening, a non-urgent failure alarm in the rod control system occurred.

All rods were reinserted and IRC personnel diagnosed the rod control circuitry problem, and replaced a low voltage power supply.

After the replacement, rods were withdrawn again at 3:20 a.m. the following morning.

However, another rod control non-urgent failure alarm occurred, and rod withdrawal was terminated.

ISC technicians replaced a different power supply, and additional troubleshooting revealed a faulty transistor in the failure circuit detector card.

The licensee replaced the circuit card and performed additional rod control testing.

By 11:05 a.m. on

August 8, 1996, control rod withdrawal was resumed, and criticality was achieved at 11:43 a.m..

The reactor was on-line and at full power by 1:52 p.m. on August 9, 1996.

C.

Conclusions The licensee's actions to shut down the reactor to upgrade the MOV-852A 5. B actuators were appropriate.

Despite the licensee's intention to remove the requirements for the auto-start function in Mode 2 from the ITS, continued auto-start events indicated a lack of operator knowledge of the system start logic, and of proper equipment configuration control while the AFW auto-start function is enabled.

The first AFW auto-start event was also an example of weak work and configuration control that the licensee has experienced recently during forced outages (see inspection reports 50-244/96-01; 96-05; and 96-06).

The shift supervisor approved the control valve test, and the work package for the test was not sufficiently detailed to permit a sufficient understanding its full impact on other plant equipment.

Safe't In ection S stem Accumulator Leaka e

a 0 Ins ection Sco e (71707)

For most of the inspection period, the safety injection (Sl) accumulators experienced leakage and had to be periodically refilled. The inspectors observed the licensee's response to this leakage and their evaluations'of its operational impact.

b.

Observations and Findin s During RCS pressurization and heatup following repairs to MOV-852A and MOV-852B, the Sl accumulator test line relief valve RV-887 lifted when Sl system check valve backleakage from the RCS caused Sl system pressure to exceed the relief valve set point. "The'licensee made several attempts to seat the check valves by "bumping" them with short Sl pump starts and by venting the Sl discharge header through the pump recirculation line. Eventually, the Sl check valves were successfully seated; however, RV-887 failed to reseat.

The Sl accumulators leaked water through one or more leaking Sl system valves and the leaking RV-887. The leakage was initially small ((0.2 gpm) and required Sl pump operation about once per day to maintain the accumulator level within the ITS band.

By the end of the inspection period, the leakage grew progressively larger (60.75 gpm) and required two to three SI pump starts per day.

All RV-887 leakage was directed to and collected in the pressurizer relief tank.

Replacing the RV would require the plant to be in cold shutdown since the valve is not isolable with'out declaring the Sl system inoperable.

The licensee used acoustic monitoring during various attempts to identify and isolate the leak paths in the Sl system.

Two manual valves (V-880B and V-880C) were shut to isolate four upstream air operated valves that were suspected of leaking.

Acoustic monitoring also indicated that the B-accumulator fillvalve was leaking.

However, RV-887

continued to leak throughout the remainder of the inspection period, and the maximum leakage measured on August 24, 1996 was 0.75 gpm.

The PORC evaluated the leakage when it reached approximately 0.5 gpm and decided that a maximum limit of 1.0 gpm could be allowed before further evaluation or action would be needed.

Operators were requested to inform management if leakage reached 1.0 gpm; however, no time limit was placed on plant operation if the leakage reached that level. At that time, the licensee considered that the leakage was not excessive, and that it was not necessary to shut down the plant to repair the leakage.

However, the licensee considered that evaluations should be performed to identify the specific leak paths and to attempt to rotate the valve stems of valves suspected of leaking in an effort to seat the valves.

The leakage required frequent operation of Sl pumps to maintain the level in the Sl accumulators within the band required by TS.

Control room operators alternated between the B-Sl and C-Sl pumps for the refills and kept detailed logs of the amount of water supplied to the accumulators.

The operators then compared those values to the level changes in the PRT and confirmed that all of the leakage was internal to the Sl system, and was not being lost outside the system.

Several operators stated that the frequent Sl pump starts were not a burden as long as the frequency remained less than once or twice per shift (approximately 15 minutes for each run).

However, the licensee planned to install a temporary modification pump dedicated to restoring accumulator level.

Some operators expected that an operational burden may be imposed if a temporary modification pump was installed because its lower capacity would require additional time and personnel to restore the accumulator levels.

Near the end of reporting period, all efforts to identify the specific leak paths and to stop the leakage were not successful.

The leakage was continuing to increase and the Sl pump starts were becoming more frequent.

The frequent Sl pump starts became a concern to the NRC, and a conference call with RGSE management was conducted on August 22nd to discuss the extent of the leakage and to review its potential safety consequences.

As noted above, the licensee also expressed concerns about the frequent pump starts, and developed a plan to install a high pressure positive displacement pump as a temporary modification dedicated to maintaining accumulator level that would relieve the frequent Sl pump starts.

The pump would be installed in the auxiliary building and was designed to take a suction from the RWST and discharge into the Sl header where flow could be diverted to each accumulator fillline. A temporary modification design package and safety evaluation for the pump were near completion at the end of the inspection.

The licensee also planned to perform additional acoustic monitoring identify all leaking Sl valves.

Conclusions The inspectors considered that the licensee's decision to establish a leak limit of 1.0 gpm was appropriate; however, it was not clear what action would be taken once the limit was reached, or how long operation would continue with the leakage less

than 1.0 gpm.

The licensee continued to analyze, monitor and troubleshoot the Sl system accumulator leakage up to the end of the inspection period (i.e. for approximately three weeks).

The NRC was concerned about the Sl accumulator leakage, its impact on operators and system reliability, and concluded the licensee's corrective actions to address the issue were not aggressive.

02.4 Feedwater Block Valve V-3984 Leaka e

ar Ins ection Sco e (71707)

In June 1996, shortly after startup from the refueling outage, a small steam leak developed in the bonnet area of feedwater block valve V-3984. The inspectors reviewed the licensee's efforts to repair it as it became progressively worse (see also sections M1.1.2 and E2.2 below).

b.

Observations and Findin s Valve V-3984 is a 14 inch manual isolation valve on the inlet side of the B-feedwater regulating valve (B-FRV), and is normally open during plant operations.

The valve and it's adjacent piping were designed and maintained under ANSI pipe design code B31.1, "Power Piping," and are classified by RG&E as "safety significant" since they provide part of the pressure boundary in the water supply to the steam generators.

The initial steam leak in the valve bonnet area was small and consisted of a very light steam flow from a single spot near the bonnet.

Maintenance technicians applied a maximum torque of 600 foot-pounds to the bonnet bolts, but the leak continued.

During the period from August 12 to 13, 1996, a marked increase in the leakage occurred to where a higher steam flow was emanating from 12 preexisting holes in the valve body.

The steam flow was high enough to make a noticeable increase in air temperature within approximately 10 feet from the valve; however, the increased leakage did not initially appear to represent an operational problem.

As the leakage became progressively worse over the course of several days, it

~

appeared to have potential personnel safety implications, and the potential for becoming significantly worse could not be very well evaluated, since the exact leak path was not known. After the reactor tripped on a feedwater isolation on August 20, 1996, a transient high pressure condition occurred in V-3984 since both main feed pumps continued to operate.

After the B-FRV shut, the valve seal leakage appeared to increase slightly.

The licensee prepared to perform an on-line leak repair of the valve using a seal injection technique after the reactor returned to power.

However, the first repair attempt was not successful and the licensee made preparations for a second attempt that would require a modification by drilling into the valve body in a pressure boundary area.

On August 22, 1996, a conference call between NRC and RGSE was held to review the plant and personnel safety implications of the leakage and the engineering evaluation supporting the second on-line repair.

The personnel

safety concerns appeared to be appropriately addressed.

However, the licensee had not yet formally evaluated the potential for steam cutting of the valve body or bolts; and had not yet conducted a formal stress analysis of the injection forces inside the seal area of the valve.

Subsequent analysis demonstrated that steam cutting of bolts and drilling into the valve pressure boundary would not significantly jeopardize the integrity of the valve body or its pressure boundary.

The seal injection pressures were shown to be within the structural capability of the valve and bonnet/seal assembly.

At the end of the inspection period, the licensee determined that the second seal injection attempt should not be performed due to the uncertainty that further seal injection would be successful, and planned to defer the valve repair for a planned outage when it could be disassembled, inspected, and repaired.

C.

Conclusions 025 Once the valve leakage increased significantly, the licensee made adequate preparations for the repair. After the first repair was not successful, the licensee's engineering evaluations for potential steam cutting, and for structural and pressure boundary integrity of the valve were thorough and timely, The licensee's decision to defer the repairs to an outage were appropriately based upon the uncertainty of success of an additional injection repair.

Automatic Reactor Tri Due to Low Steam Generator Level Followin Isolation of the B-Feedwater Train

Ins ection Sco e (71707)

On August 20, 1996, at approximately 2:46 p.m., the reactor automatically tripped after a low-low level occurred in the B-steam generator (B-SG). The low level was caused by an unexpected closure of its feedwater regulating valve (B-FRV), after a fault occurred in a differential pressure transducer that is part of the valve controller.

The inspectors observed the control room operators'esponse to this event and the licensee's response to other equipment problems following the trip.

b.

Observations and Findin s At 2:42 p.m. on August 20, 1996, with the plant at 100% power, the 8-FRV controller received a failed input signal that caused the valve to go full closed.

Operators attempted to restore full feedwater flow to the B-SG, but were not able to recover flow before SG level decreased to the low level trip point.

The B-FRV and A-FRV controllers automatically transitioned to the manual mode during this event.

The A-FRV received a full open demand signal and consequently went full open, causing the A-SG level to increase while the B-SG level was decreasing.

When the B-SG level reached 20%, the control room foreman ordered a manual reactor trip, but level reached the automatic trip setpoint (17%) first, and the reactor tripped at 2:46 p.m..

The A-SG level continued to increase and operators manually closed the A-FRV, but level continued to increase due to the addition of

AFW that eventually caused the A-SG level to reach its high level trip setpoint.

However, this had no consequence for plant operators or equipment because the A-FRV was already shut.

Operators controlled AFW and returned level in both SGs to normal.

Following the reactor trip, operators exercised the Emergency Operating Procedures (EOPs), first entering E-O, "Reactor Trip or Safety Injection", and subsequently transitioning to ES-0.1, "Reactor Trip Response".

Once the provisions of that EOP was satisfied, the operators maintained the plant in Mode 3 per operating procedure 0-3, "Hot Shutdown With Xenon Present."

Shortly after the reactor tripped, the licensee determined that the immediate cause was a loss of electrical continuity in a transducer (IP-476) that provides a current to pressure (I/P) control signal to the automatic digital feedwater control system (ADFCS) for regulating the valve position.

The loss of continuity created a "no signal" condition within the ADFCS logic, which then caused the FRV to close.

Upon investigation, it became immediately apparent that a screw was missing from the terminal block inside the transducer that connects the terminal block with a terminal block connection board, and assures electrical contact is maintained with the internal wires.

The licensee performed repairs to the transducer and inspected others of similar design (Rosemount Model 3311).

One other transducer in the feedwater control system was also missing a screw in its terminal block, but proper electrical circuit continuity was still maintained.

Another transducer associated with the A-ARVhad a broken screw installed.

The licensee initiated a human performance evaluation to identify the root causes leading to the reactor trip. Aside from the missing screw, other questions centered around potential vibration of the transducer, tampering, and maintenance work practices that could have caused the terminal block to lose continuity with the attached board.

High vibration was eliminated as the probable cause since the transducer is solidly mounted to structural steel in the turbine building.

The licensee determined that a previously allowed work practice accounted for the missing screw, and that the reactor trip was a maintenance preventable functional failure.

Immediately following the reactor trip, auxiliary feedwater discharge throttle valve MOV-4008 did not throttle back to a 200 gpm as expected (the actual flow was 180 gpm) for Mode 3. The licensee did not declare this valve inoperable, even though the basis for ITS LCO 3.7.5, "AuxiliaryFeedwater System," assumes that each motor-driven AFW train is capable of supplying its respective SG with R 200 gpm total flow while in Modes 1, 2, and 3.

Emergency operating procedure (EOP)

E-O, "Reactor Trip or Safety Injection," Step 15, in use immediately after the reactor trip, requires plant operators to verify that "total AFW flow is greater than 200 gpm."

Procedure E-0 also states that valves MOV-4007 and MOV-4008 will throttle to provide the minimum for decay heat removal.

The shift supervisor on-shift during the reactor trip indicated that these requirements were verified because both trains were supplying a total AFW flow M 200 gp When MOV-4007 and MOV-4008 were replaced during the last refueling outage, engineering treated the changeout as an "equivalency" since the nominal valve size and hydraulic characteristics appeared to be the same in the throttled flow region.

However, the replacement valves had a plug design that made throttle flow much more sensitive to stem position.

The correct stem position for these valves is also effected significantly by SG pressure and valve differential pressure when the actuator motor is tripped.

Consequently, several instances have occurred since the refueling outage where accurate and repeatable throttle flows were not achieved.

The unreliable throttling of these valves has also been exacerbated because an apparent design deficiency (see NRC inspection report 50-244/96-06, sections 02.2, E.1.1. and M1.2). The licensee contacted the valve manufacturer to determine if the plugs removed with the old valves could be used in the new valves since they appeared to be better suited to proper throttling in these type valves.

The manufacturer indicated that the plugs were interchangeable and were equivalent from a functional standpoint.

The licensee indicated that plans would be made to replace the existing plugs with the old design.

On August 4, 1996 the Nuclear Safety and Licensing (NS&L) Department issued a

letter to the plant Shift Supervisors indicating that a minimum nominal auxiliary feedwater flow of 200 gpm per pump without manual operator action is assumed in the accident analyses for the following transients:

1) Loss of Main Feedwater, 2)

Small Break LOCA, and 3) SG Tube Ruptures.

These transients are analyzed with SG pressures from 1005 to 1096 psig.

The letter recommended that all adjustments in AFW flow should be made at hot zero power conditions (Mode 3) so that the proper throttle range will be automatically achieved.

On August 5, 1996, the PORC approved a safety evaluation (SEV-1073, Rev.0) which stated that setting the control logic for MOV-4007 and MOV-4008 in Modes 2 and 3 may yield 5 200 gpm during full power conditions.

It also stated that a worst case loss of MFW transient at power assumed a MFW line break inside the intermediate building that causes a loss of both primary AFW pumps.

In this case, credit is taken for operators to place the standby AFW system in service within 10 minutes.

However, if the break occurred outside the intermediate building, the accident analysis assumed that 200 gpm is available from the primary AFW pump supplying water to the faulted SG.

The inspector questioned the operability of both valves when they failed to throttle in the 200- 230 gpm range as stated in the ITS bases and SEV-1073, Rev.0, following a loss of main feedwater, since that transient alone will not generate a

reactor trip at the Ginna Station.

NS&L personnel stated that the accident analysis credits the loss of 30 gpm to account for an AFW recirculation valve failing open

~ and diverting this flow away from the SG.

Also, no accident analysis assumes that AFW flow is available prior to a reactor trip.

RG&E Engineering subsequently ran several computerized accident scenarios assuming that no AFW flow was available for 10 minutes after a reactor trip from Mode 1.

In all cases, the SGs did not achieve a dryout condition prior to 10 minutes.

NS&L considered the existing procedural guidance for operator action in one minute to establish a 200 gpm total AFW flow was sufficient for AFW

operability.

The licensee stated that these results and additional considerations for AFW accident requirements will be clarified in a future revision to SEV-1073, and in a revision to the ITS bases for AFW.

The inspector considered the analytical results were adequate for the subsequent reactor startup; however, pending NRC review of the revised SEV and the ITS bases, this item is unresolved (URI 50-244/96-07-02).

After the reactor trip on August 20, 1996, the intermediate range nuclear instrument N-36 indication did not decay below the setpoint required to automatically energize the source range detector (5 X 10" amps).

Both intermediate range channels must to be below 5 X 10" amps to automatically energize the source range instruments; however, the source range instruments were manually energized by the operators as required by procedure ES-0.1, Step 13 (P-6 permissive defeat on the MCB). The inspectors confirmed that none of the automatic trip functions were affected by the high reading in the low range and no TS inoperability condition existed at that level with the plant in Mode 3 after the reactor trip.

Westinghouse had previously informed RGKE that it may be necessary to wait approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after shutdown before adjusting the compensating voltage on the intermediate range instruments due to local gamma energy distribution changes after a reactor shutdown.

Since the reactor was started up prior to this time frame the compensating voltage for N-36 was not adjusted during this shutdown, nor was it required to be adjusted.

During the subsequent power ascension, both intermediate range detectors were in close agreement when the instruments reached 10" amps.

The intermediate instruments are beyond the level affected by compensating voltage at 10'mps.

After repairs were made to the B-FRV pressure transducer IP-476, the reactor was returned to power operations the following day.

The reactor was below Mode 1 for approximately 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />.

The licensee elected not to repair the Sl accumulator leakage, the FW isolation valve leakage, or to change the valve plugs on MOV-4007 and MOV-4008 with the previous design.

Instead, the reactor was online and at full power by 5:33 a.m. on August 21, 1996.

Conclusions The inspectors determined that control room personnel appropriately responded to the August 20, 1996 reactor trip.

EOPs were correctly implemented, and the transition to normal operating procedures to maintain Mode 3 operations was performed efficiently. The inspectors agreed with the licensee's conclusion that the reactor trip was a maintenance preventable functional failure.

The equipment problems that occurred after the reactor trip were adequately resolved prior to the plant startup.

However, the equipment problems that occurred before the reactor trip were not addressed during the shutdown period.

RGSE's decision to return to power operations within 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> after the trip with several

'

degraded equipment conditions (some in safety-related equipment), lacked the conservative operating philosophy usually noted.

Operator Knowledge and Performance 04.1 Failure of The A-SW Pum to Start a.

Ins ection Sco e (71707)

The A-service water (A-SW) pump failed to start during a routine swap of the operating pumps from the main control room.

The inspector reviewed the licensee's follow-up actions to this event.

b.

Observations and Findin s At 3:56 p.m. on August 2, 1996, during a service water pump swapover, the A-SW pump failed to start upon closing its MCB switch.

Operators received a white light when the pump switch was placed in start, indicating that the breaker did not close upon demand.

An AO was sent to investigate the pump start failure. The breaker was not fully racked in and was out of the locked position.

The AO re-racked the breaker back in and the A-SW pump was started satisfactory.

The electrical shop performed a diagnostic test on the breaker and found it functioned satisfactorily.

The licensee initiated an ACTION Report (AR 96-0764) to evaluate this problem and eventually attributed the improperly configured breaker to operator error.

For corrective action, the licensee planned to perform training for auxiliary operators to reinforce the techniques for proper rackout and locking of electrical breakers.

Consequently, a training work request (TWR 96-0950) was initiated which requested training for AOs on proper breaker operation.

The training department has reviewed the ACTION Report and will provide classroom training on the proper breaker racking and latching techniques.

Since the training work request was submitted during the middle of the current maintenance training cycle, it will be scheduled for the next cycle which starts October 21, 1996.

c.

Conclusions The inspector concluded that the licensee promptly diagnosed the cause of the pump start failure and corrected the condition.

Training for all AOs on breaker operation is appropriate.

/

04.2 AFW Pum Automatic Start After Latchin the Main Turbine for Control Valve

~Testin a.

Ins ection Sco e (71707)

The inspector reviewed the circumstances associated with an automatic start of the AFW system on August 6, 199 '

b.

Observations and Findin s

As discussed in section 02.2, an AFW auto-start occurred during a main turbine control valve (CV-2) test.

The inspectors determined the operators did not sufficiently understand the scope or effect of the test on plant systems, and did not anticipate that latching the main turbine with both MFP breakers open would cause an AFW auto-start.

The initial conditions identified in maintenance procedure M-109 were also vague as to equipment configuration requirements.

The licensee has recently experienced work and configuration control problems during forced outages.

Despite recent attempts to correct this problem, this event represented another example of the problem where corrective actions have not been fully effective.

In this case, the shift supervisor approved the CV-2 test without understanding its full impact on other plant equipment.

As a result of the incident, the operations manager directed that all future work planned during the outage be approved through the operations office staff.

C.

Conclusions The inspectors noted that the operators have been challenged on several recent

~

occasions by unanticipated automatic starts of the AFW system while in Modes 2, 3, and 4. These events have been associated with required AFW availability in these modes as imposed by the ITS. The inspector determined that work control, configuration control, and operator knowledge have been challenged since the ITS were implemented in February 1996.

Miscellaneous Operations Issues 08.1 Closed LER 96-008: Main Feedwater Pum Breakers 0 en Due to Low Seal Water Differential Pressure Results in Automatic Start of the A-Auxiliar Feedwater A-AFW Pum LER 96-008 reported an automatic start of the B-AFW pump (ESF actuation) during a plant heatup on July 7, 1996, when the condensate system was being realigned from the recirculation cleanup mode to the normal operating configuration.

During the course of the system realignment, a low differential pressure condition in the A-MFP shaft seal occurred that caused the A-MFP breaker to open.

At the time, the A-MFP breaker was racked out in the test position to simulate a closed position and to satisfy technical specification requirements to have the automatic start of the AFW pumps available while in Mode 2. Since the B-MFP breaker was already open, the two open MFP breakers provided the necessary logic conditions to generate a

start signal to both motor-driven AFW pumps.

The A-AFW pump was already running, and only the B-AFW pump started.

No operational transients or safety consequences resulted from the event.

The event is also described in NRC inspection report 50-244/96-06.

The LER identified that during recirculation of the condensate system, there is condensate flow to the condenser hotwell that provides a differential pressure

between the condensate pump and the MFW pump suction.

When realigning the condensate system to the normal power configuration without MFW pump flow, the differential pressure may decrease to zero.

The differential pressure in this case decreased below the 15 psid alarm setpoint, but the operators were not required, and took no action under the alarm response procedure because no MFP was operating.

The LER primarily indicated that the corrective actions for this event would be to submit a technical specification change.

This will eliminate the need for automatic start of the AFW pumps while in Mode 2 and to have a MFW pump breaker closed in the test position during Mode 2 conditions.

The licensee considered the technical specification change necessary because AFW pumps are used during plant startup below Mode 1 conditions before a MFW pump is started, and AFW is always available above Mode 4.

After submitting the LER, the licensee made a temporary change to startup procedure 01.2, "Plant Startup From Hot Shutdown to Full Load," to ensure that the condensate system is reconfigured out of the recirculation mode before putting a MFP breaker in the test position.

The inspector considered that the LER adequately described this event and properly addressed the root causes.

Elimination of the ITS requirement for the auto-start feature is a planned future licensing action.

Changing the startup procedures to preclude a recurrence was appropriate.

This LER is closed.

08.2 0 en LER 96-009: Leak Outside Containment Due to Weld Defect Results in Leak Rate Greater Than Pro ram Limit On August 22, 1996, the licensee submitted LER 96-009 to report an unisolable leak in containment spray system piping outside containment that exceeded a

program limitof 2 gallons per hour. (See report section 02.1 above).

LER 96-009 reported containment spray (CS) system leakage above a program limit while the plant was operating at full power.

The "Primary Coolant Sources Outside Containment Program" is required by the improved technical specification (ITS)

section 5.5.2, but the ITS does not specify the program limit. Based upon inspection results stated in paragraph 02.1 above, it is apparent that the "program" limitof 2 gph is derived from a design basis analysis, and is also contained in the Ginna UFSAR.

LER 96-009 reported the CS system leakage as being outside the analyzed value.

The existing program limit was based upon an accident analysis which concluded that leakage below 2 gph would keep post accident exposure levels at the low population zone boundary below 10 CFR Part 100 limits following a design basis LOCA. The LER concluded that the measured leakage was not outside the design basis limits for ECCS leakage.

Preliminary results of a subsequent Westinghouse analysis imply that leakage in this specific location can be K 30 gph.

The LER indicated that a supplement would be issued by September 20, 1996.

However,

pending NRC revieW of the supplemental LER, and the resolution of unresolved item URI 50-244/96-07-01, this LER will remain open.

II. Maintenance M1 Conduct of Maintenance M1.1 Maintenance Observations The inspectors observed portions of plant maintenance activities to verify that the correct parts and tools were utilized, the applicable industry code and technical specification requirements were satisfied, adequate measures were in place to ensure personnel safety and prevent damage to plant structures, systems, and components, and to ensure that equipment operability was verified upon completion of post maintenance testing.

The inspectors observed portions of the following maintenance activities:

M1,1.1 Installation of New Motors and Cables for MOV-852A and MOV-852B a.

Ins ection Sco e (62703)

On August 2, 1996, at 8:00 p.m., following several discussions with NRR and NRC Region I, the licensee initiated a voluntary plant shutdown because of inadequate safety margin on the motor operated core deluge valves MOV-852A and MOV-852B. The inspectors witnessed portions of the maintenance activities that installed upgraded MOV actuator motors and power cabling.

b.

Observations and Findin s Au ust 3 1996 The inspector conducted walkdowns of cable trays in the intermediate building and the containment building during the installation of the'new cables.

The inspector noted plant department managers and senior RG&E technical personnel were also in the field performing walkdowns.

The licensee determined that MCC-C and MCC-D in the intermediate building would be upgraded with larger size breakers to accommodate the larger size motors for MOV-852A and MOV-852B, respectively.

There were several scheduling meetings held with management, engineering and senior technical personnel to discuss installation of equipment i.e., cables, motors, splicing, motor operator reduction gears, testing, materials, equipment qualification, seismic qualification, and scheduling changes.

The licensee also developed a time line for installation of equipment and required testing.

Au ust 4 1996 The inspector conducted walkdowns in the intermediate and containment buildings to evaluate the cable installation.

All cabling was installed except for the last 30 ft

'Qs

to each of the motors inside containment.

The actuators were modified by installing larger capacity..motors, breakers and power cables, and by replacing the actuator internal gearing to achieve a higher torque/thrust output, RGRE management and engineering held meetings to discuss EQ qualification and seismic support requirements for the 30 ft portion of the containment motor cables which will be put in conduit the next refuelling outage.

Engineering performed a calculation and found that the temporary seismic installation was acceptable.

The inspector also witnessed EQ splices inside containment, at the containment penetrations, and at the motors.

Diagnostic testing of motor actuators was completed satisfactorily.

c.

Conclusions The inspector concluded that there was good communication between management, engineering and the technical staff. All issues were thoroughly discussed in a professional manner.

Electrical Engineering management from the corporate office was represented onsite during the course of this activity, and senior engineers were onsite and available to resolve questions and problems from the field staff. This approach led to quick and timely resolution of questions and field problems during installation.

The RG5E electrical maintenance staff demonstrated proficiency and expertise during the installation of the new motors, cabling and breakers.

M1.1.2 On-line Leak Re air of Main Feedwater Valve V-3984 a.

Ins ection Sco e 62703 The inspectors observed the licensee's efforts to perform an on-line leak repair of the Main Feedwater (MFW) block valve V-3984 on August 21, 1996.

b.

Observation and Findin s In preparation for the leak repair activities, the licensee roped off a wide area around the valve to restrict access.

The contractor installed 12 small seal injection valves into existing seal ring "knock-out" holes near the bonnet of the valve, and injected sealant into the valve's bonnet seal region.

The injection proqess caused a larger steam plume to be directed toward the south wall of the turbine building and onto a panel about 6 feet from V-3984 that had small solenoid pilot valves attached.

It also directed steam toward the B-FRV bypass valve.

The licensee took precautions to cover these valves after the steam flow increased significantly and impacted onto

'he panel and valves directly.

The initial seal injection repair was not successful in stopping the valve seal leak, and the licensee postponed additional repair activities until engineering could evaluate the methodology for attempting further repair '

c.

Conclusions

Although the licensee did not take the opportunity to disassemble V-3984 after the reactor tripped on August 20,.1996, the inspectors concluded that maintenance activities for the subsequent on-line leak repair were adequately controlled.

The licensee and contractor personnel properly adhered to maintenance and safety procedures during the initial repair attempt.

The licensee appropriately avoided going beyond the initial repair so that additional evaluations could be made for further repairs.

M1.1.3 Installation Testin of New Service Water SW Pum Vacuum Breaker and Motor a.

Ins ection Sco e 62703 The inspectors observed the installation and testing of the A-SW mechanical vacuum breaker, and three replacement motors.

b.

Observations and Findin s

As part of the recent installation of a rebuilt A-SW pump, the licensee replaced the old pump discharge check valve with a new low leakage nozzle-type check valve.

The old check valves had previously created a vacuum condition in the pump casing when the SW.pump was shut-off and the check valve seated, The vacuum caused the pump packing to tighten, and caused premature wear on the pump shaft.

The licensee also added a mechanical vacuum breaker to each SW pump discharge to eliminate the vacuum formation and prevent premature shaft wear.

The inspectors observed the installation of a new mechanical vacuum breaker on the A-SW pump.

The A-SW replacement motor was a commercial-grade procurement item (manufactured by U.S. Motors), and the licensee performed the following safety-related dedication tests in accordance with procedure CGIEE-96-006, "Commercial Grade Item Engineering Evaluation."

Procedure SM-95-075.1, "Service Water Pump 1A Motor Replacement Modification," was used to perform installation, static and functional testing, and the subsequent return to service.

The inspectors observed portions of the following motor testing:

~ Resistance testing of the motor windings;

~ Hipotronics" testing using a variable power supply transformer was conducted to verify motor voltage and amperage parameters at low voltage starts;

~ Motor vibration measurements;

~ Motor bump-starting; and

~ Direct Current "hi-pot" testing and "megger" tests.

The inspectors reviewed the above test results and concluded that, with the exception of the hi-pot and megger tests, the results were satisfactory based upon the recorded data.

All measured test parameters for the first four tests noted above were within the acceptance criteria defined by the test procedure.

The hi-pot and

megger tests were not satisfactory because they showed excessive micro-amperage current leakage that was inconsistent with previous on this and other motors tested at US Motors and the Schulz Electric Co. The licensee notified the manufacture of these test results.

il After discussing the test results with U.S. Motors, the licensee dissembled the motor for an internal inspection and observed the following conditions:

~ The motor shaft threads and shaft lock nut were damaged.

~ Baked insulation ties that hold the winding motor leads were cut and there was a dark spot on the winding lead, probably due to excessive heat.

~ There were metal shavings on machined parts and in the motor oil basin.

~ Unidentified debris was present around the motor lead-in windings.

The licensee took laboratory samples of the debris around the motor lead-end windings and sent the material off-site for laboratory analysis.

The laboratory results for the debris present on the inside windings on the lead-end of the stator suggested a strong potassium peak with small oxygen, aluminum, silicon, titanium, iron, and copper peaks.

The results for the debris present on the outside windings on the lead-end of the stator suggested that the debris is primarily organic, with a strong chlorine peak.

A strong silicon peak was also present in discrete particles contained in the debris.

The debris on the inside windings appeared to originate from the flex residue at the brazed motor joints which came off the rotor during motor operation and collected on the stator windings.

The licensee replaced the existing motor with a second new motor.

Megger testing was conducted on the motor prior to installation.

However, the readings showed fluctuating current leakage which was inconsistent with previous data from US Motors and Schulz Electric. The licensee assumed that the second new motor had the same problems as the first motor which was found to be unsatisfactory.

The second new motor was dissembled in the presence of the manufacturer's representative.

The licensee observed the following conditions:

~ The motor shaft threads and shaft lock nut were damaged;

~ There were metal shavings on machined parts and in the oil;

~ The motor lead insulation was tom near the motor windings; and,

~ The motor's electrical leads had been scraped, causing a loss of insulation on the motor leads.

This was apparently due to excessive buildup of varnish that had been removed.

The manufacturer's representative reviewed all data from the first and second A-SW motors and determined that the defects had occurred during manufacturing.

The representative discussed the motor defects with the parent company, who agreed to have all necessary repairs completed at a factory authorized repair facility (Schulz Electric Co.).

The licensee sent one of its electrical engineers to verify and document all corrective work done on the motors at the repair facility. The licensee also had in place a procedure to dedicate the commercial-grade motors at the Schulz

Electric facility, and then to complete the onsite dedication for safety-related use using RGRE's dedication program during installation and return to service.

The licensee then replaced the second new motor with a third new motor.

Electrical testing performed on that motor did not identify any problems. However, the motor started to experience vibration problems that were in excess of the licensee's alert range.

In an attempt to reduce vibration levels, 0.25 inch shims between the motor and pump pedestal were removed and the pump was run again.

The vibration readings show that the spectrums and waveforms were similar, but overall values decreased from 0.6955 in/sec to 0.4013 in/sec.

The motor was again removed from the pump pedestal and paint was removed that kept the two surfaces from fully mating.

Vibration readings were then taken showing that spectrums and waveforms had reduced to 0.1371 in/sec and resulted in the motor running much smoother and below the alert range.

Vibration readings were again taken after approximately twelve hours of run time after the overall levels stabilized.

Even though the diagnostic testing was within acceptable limits, the licensee intended to treat the motor deficiencies as a common cause failure mechanism for all new SW motors, (i,e., defective manufacturing),

and continued to do electrical diagnostic and vibration testing to identify any further anomalies.

As a result of the discrepancies identified above, the licensee initiated the following action plan for the ongoing SW pump motor replacements:

~ Upon return of the first motor from Schulz Electric, the existing motor on the A-SW Pump will be replaced.

Testing will be performed on the repaired motor in accordance with existing procedures to ensure acceptable performance prior to it being declared operable;

~ The motor removed from A-SW pump will be sent to Schulz Electric to be inspected, tested and repaired to ensure acceptable performance;

~ Upon return of the second motor from Schulz Electric, the existing motor on D-SW Pump will be replaced.

The motor removed will be disassembled and the oil bath cleaned.

Also, the motor windings and motor leads will be tested and inspected to ensure acceptable motor performance; and,

~ After repairs are performed on the two remaining 350 HP spare motors (i.e.,

motors previously installed on A-and D-SW pumps), they will be used to replace the 300 HP motors on the B-and C-SW pumps.

/

The licensee also initiated a 10 CFR Part 21 review for reportability.

Replacement of all SW motors will require the licensee to revise UFSAR Sections 9.2.1.2.1 and 3.5.1.3.7 to reflect the change from 300 to 350 HP.

c.

Conclusions Installation and testing of the A-SW pump motors was effectively performed in a thorough and well coordinated manner.

The licensee's electrical engineering and maintenance technicians maintained a questioning attitude; and the technical involvement of engineering and management led to the timely resolution of as-

i

received defective conditions, and subsequent motor vibration problems.

There was good coordination between maintenance, engineering, and management in resolving the identified deficiencies.

M1.1.4 Emer enc Diesel Generator EDG Dia nostic Testin a.

Ins ection Sco e 62707 The inspectors observed diagnostic testing of both EDG's.

b.

Observations and Findin s The inspectors witnessed diagnostic testing and evaluation on A-and B-EDG by the licensee's contractor (CARMA). The contractor performed a pre-outage analyses and provided corrective action recommendations which consisted of the following:

A-EDG

~ Fuel Injector Nozzle replacements:

Cylinders 2L and 4L

~ Valve Clearances:

Cylinders 1L, 6L, 7L, 1R, 2R, and 6R

~ Boroscopic Inspection concerns:

Cylinder Liners 3R and 4R B-EDG

~ Fuel Injector Nozzle replacements:

Cylinders 1L, 3L, 8L, 2R, 4R, 5R, 6R and 7R

~ Valve Clearances:

Cylinders 2L, 3L, 4L, 6L, 8L, 2R, 3R, and 8R

~ Boroscopic Inspection concerns:

Cylinder Liner 4R The contractor returned on July 24, 1996, to perform the diagnostic test evaluation on each EDG after the licensee performed the recommended changes during the 1996 refueling outage.

Preliminary tests results for both A-and B-EDGs were satisfactory.

A final report was scheduled to be issued within forty-five days after completing the tests.

The report will be review by RGSE and a copy sent to the ALCO Owners Group (OG) as input into the development of a revised EDG owner's manual.

The OG and the manufacturer are revising their maintenance procedures to accommodate a change in maintenance philosophy.

RGSE is part of ALCO OG, which is collectively addressing technical concerns related to the operation, testing, and maintenance of ALCO diesels.

Some of the specific issues being addressed by the OG are as follows:

~ Root cause analyses of reliability issues e.g., fractured pistons, cracked heads, seizure of fuel injection pumps.

~ Changing the approach of engine maintenance from "Time-based" to "Condition-based" maintenance.

~ Provide improved guidance e.g., revise engine inspection manual, develop lube oil and jacket water guidance manual, and develop engine analyzer guidance manual.

The owner's group for ALCO diesels is developing a revised maintenance manual that is more appropriate for emergency diesels used in nuclear standby service.

The revised maintenance manual was designed to provide new guidance to clarify

requirements regarding "time-based" engine tear-downs.

Included in this effort would be the diesel vender endorsement and an exchange of information with the NRC.

c.

Conclusions The inspectors concluded that the scope and performance of the diagnostic testing on both the A-5 8-EDGs was in accordance with the recommendations from the owner's group consultant.

The licensee is taking a proactive approach in assisting the owner's group develop new guidance that will be provided in an upgraded maintenance program for ALCO EDGs.

M1 ~ 1.5 Emer enc Diesel Generator EDG Testin and Troubleshootin a.

Ins ection Sco e 62703 The inspectors observed post-maintenance testing (PMT) and troubleshooting on the B-EDG's motor operated potentiometer (MOP) portion of the Basler regulator.

b.

Observations and Findin s The inspectors witnessed maintenance testing and troubleshooting on the B-EDG.

The licensee conducted voltage testing and connected recording instrumentation to terminals 14 and 15 of the motor operated potentiometer (MOP) portion of the Basler regulator to monitor and record Kw and voltage while the EDG was being loaded and unloaded.

The licensee had previously detected a problem with the EDG dropping load by 100 Kw. During testing it was determined that there was a voltage dip on MOP terminals 14 and 15 prior to the EDG output and frequency oscillations.

This occurred at about 900 kw output.

Based on a review of the test data it was determined that the potentiometer was unreliable while in the test mode.

However, the licensee determined that there were no diesel operability concerns with this condition.

The manufacturer reviewed the licensee's test data and made the same conclusion.

The licensee was in the process of replacing the potentiometer at the end of this inspection.

c.

Conclusion The inspectors concluded that the observed Results and Test personnel exhibited good awareness of a power anomaly observed during shutdown from a surveillance of the B-EDG. Technicians demonstrated good skills in the use of special test equipment during subsequent testing and troubleshooting to diagnose the problem.

The testing was well coordinated and personnel properly adhered to the special test procedure.

The problem was properly diagnosed and appropriate corrective actions were take '

M1

~ 1.6 Installation of Meters on the Instrument Buses And Twinco Panels Ins ection Sco e 62703 The inspectors reviewed the installation of digital voltage meters on the Instrument buses and the Twinco panels.

b.

Observations and Findin s The licensee conducted a planning meeting to review work procedures and installation requirements for the digital voltage meters on the Instrument buses and the Twinco panels.

Prior to installation, each meter was calibrated and bench tested to assure the meters performed their function as designed.

As part of the work order, the licensee removed the existing sliding link decks and replaced them with isolation fuse decks for both the Instrument buses and Twinco panel voltage meter enclosures.

This provided meter electrical isolation from the safety-related buses by allowing the technician to pull the isolation fuses within the meter enclosure.

The meters are also designed to indicate bus voltage, notwithstanding harmonic distortion from the inverters, with an accuracy necessary for satisfying the ITS requirements.

The inspectors observed the installation of the Instrument buses and Twinco panels digital meters in the control room and relay room. A QC inspector was present during the installation and no deficiencies were noted.

C.

Conclusion Safety and work procedures were reviewed by the engineer and technicians prior to the start of work. The inspectors concluded that the scope of work identified in the procedure provided sufficient technical guidance for the technicians to properly.

install the digital voltage meters.

The installation and testing were well coordinated and performed in a professional manner and in accordance with applicable codes and standards.

These included ANSI/IEEE-498-1985, "Calibration and Control of Measuring Test Equipment used in Nuclear Facilities," and NEC-70-1993, "National Electric Code."

I e

M1.1.7 A-Emer enc Diesel Generator EDG Thermo ra h

Surve a e Ins ection Sco e 62703 The inspectors observed a thermography survey of the A-EDG lube oil cooler discharge piping.

The survey was a new initiative for EDG diagnostic testing and for validating preventive maintenance (PM) frequencies.

b.

Observations and Findin s The inspectors witnessed a thermography survey of the A-EDG lube oil cooler discharge piping.

The licensee was taking readings prior to EDG start and during EDG full load while lube oil was flowing through the oil cooler.

As the lube oil flows towards the piping tie, near temperature thermocouple indicator TI-2658,

downstream of the oil cooler, it mixes with oil flowing in the bypass line.

Thermography testing indicated that pipe surface temperature is warmer at the top than at the bottom when discharging from the oil cooler.

The licensee was using a Thermovision infrared imaging camera, a digital infrared scanner, and a micro-mite thermocouple readout thermometer to measure and image spot temperatures.

The licensee is providing a representative measurement of lube oil discharge temperature for use in oil cooler thermal performance monitoring.

From the data obtained, there is a small delta T, top to bottom, on the outside of the pipe.

The diesel temperature element indicator, TE-6623, seemed to be measuring the temperature at approximately the middle of the flow stream.

The licensee will repeat the thermography survey during different intervals and during the winter months when service water temperature is lower for comparison purposes and also to build a data base.

The licensee, willthen review the thermography survey data to identify any excessive high readings or undesirable trends.

In addition to the thermography survey, the components under inspection are given an external examination for any discoloration, wear, damage, or questionable noises.

During the thermography survey there were no deficiencies noted.

C.

Conclusion The licensee's initiative to develop a new test methodology using thermography appeared to provide an important improvement in the reliability centered maintenance program for the diesel generators, could improve component reliability through early detection of equipment degradation, and could validate PM frequencies.

The thermography survey was thorough and accurately documented.

M1.2 Surveillance Observations Ins ection Sco e (61726)

The inspectors observed portions of surveillances to verify proper calibration of test instrumentation, use of approved procedures, performance of work by qualified personnel, conformance to limiting conditions for operation (LCOs), and correct post-test system restoration.

The inspectors observed portions of the following surveillance tests:

II

~ PT-2.2Q, Residual Heat Removal (RHR) Quarterly Performance Test, for the A-and B-RHR pumps.

~ Calibration of the A-Sl Accumulator Level Loop Transmitters LT-938 and LT-939

~ A-Main Feedwater (MFW) Pump Seal Flow Control Valve Calibration and Test

~ Turbine Trip and Auxiliary Governor Solenoids Operability Check

~ Throttling of Auxiliary Feedwater (AFW) Valves MOV-4007 and MOV-4008

~ A-SW pump performance test

b.

Observations and Findin s While performing surveillance PT-2.20 on the B-RHR pump, the pump inboard horizontal (PIH) vibration measured value placed it in the "Alert" range, which required the frequency of the surveillance for the pump to be doubled until the cause was determined.

The licensee subsequently issued an ACTION Report (96-0809) for the B-RHR pump.

The inspector researched the history of the PIH vibration for the B-RHR pump and found it was measured in the "Satisfactory" range on the two previous surveillances performed.

Both RHR pumps passed the performance test satisfactorily.

The inspector determined that the performance test was performed satisfactorily and in accordance with plant procedures, and that the actions taken in response to the abnormal parameter measured for the B-RHR pump, including increasing the testing frequency, were appropriate.

The inspector witnessed the surveillance testing of the auto stop trip solenoid (AST), emergency trip solenoid (ET), and the two auxiliary governor solenoids (AG)

and the turbine stop valve input to the reactor protection system (RPS).

An ISC technician requested the Control Room operator to latch the turbine.

This was done to supply electrical hydraulic (EH) pressure (above 19 inches), to the control valves in order to stroke them.

The technician and the operator verified that the procedures would not make up the logic for an inadvertent actuation of the motor driven AFW pumps on a turbine trip. The technician held the turbine trip-reset lever to the reset position until the AST oil pressure increased above 60 psi (the trip set-point is 45 psi).

During the testing, the senior ISC technician properly made changes to Procedure M-51.11, Section 3.1, 3.2, to clarify initial conditions, and add a note before step 5.2.6 to reset the lockout relays after the turbine trip. The inspector noted that there were no deficiencies during this test.

The inspectors observed the throttle testing of AFW MOV-4007 and MOV-4008, to verify each MOV limited flow to 200 - 230 gpm.

The test was done four times for repeatability.

The flow control setpoints were within the requirements of ITS Basis 3.7.5.

There were no test deficiencies noted.

c.

Conclusions Overall, the inspectors concluded the surveillance activities were well controlled and coordinated.

Testing personnel properly adhered to surveillance and safety procedures.

In one case, the technicians properly changed the procedure to clarify and enhance it. The technicians displayed good technical knowledge.

Appropriate actions were taken in response to abnormal test result M4 Maintenance Staff Knowledge and Performance M4.1 Autostart of A-AFW Pum Due to MFP Breaker Not in The Test Position a.

Ins ection Sco e (62703)

The inspector reviewed the circumstances surrounding the autostart of the A-AFW pump on August 7, 1996, and interviewed the electrical maintenance shop foreman to evaluate electrical procedures for placing the MFP breakers in the test position.

b.

Observations and Findin s On August 7, 1996, with the plant shutdown in Mode 3, plant operators performing procedure 0-1.2, "Plant Startup From Hot Shutdown to Full Load," placed the AFW bypass switches on the MCB into the "Normal" position in anticipation of a transition to Mode 2 where the auto-start capability of AFW is required by the ITS.

When the switch was placed in normal, the A-AFW pump automatically started because one MFP breaker was not properly configured in the test position.

This event required a report to the NRC in accordance with 10 CFR 50.72 as a non-emergency ESF actuation.

The inspector interviewed the electrical shop foreman with regard to the procedure for putting these breakers in the test position.

Using a similar breaker (in 4KV station service bus 11B) that was out of service, the foreman demonstrated how an actuating rod for an auxiliary contact switch (52S) should be observed to verify the breaker was properly placed in the test position.

The foreman demonstrated how the breaker positioning arm traveled beyond the mechanical stop that was out of its normal position.

Consequently, the positioning arm interfered with the actuating rod for switch 52S and prevented it from signaling the breaker control circuit logic that the breaker was closed in the test position.

Electrical maintenance personnel apparently erred and did not closely observe the incorrect configuration of the breaker, the actuating rod, or the test switch when they attempted to place the breaker in the test position.

Followup training was scheduled for all electrical maintenance personnel to be completed by September 30, 1996.

c.

Conclusions The inspector concluded that the proper positioning of MFW pump breakers should be within the skill of the electrical maintenance craft and that the scheduled training to reiterate these skills was appropriat III. ~En ineerin E2 Engineering Support of Facilities and Equipment E2.1 Desi n Anal ses and Safet Review for Seismic and Pressure Boundar Qualification of Freeze Seal E ui ment Used to Isolate Containment S ra Leaka e

a.

Ins ection Sco e (37551)

The inspector reviewed the Design Analysis (DA) and Safety Evaluation (SEV)

prepared by the licensee for seismic and pressure boundary qualification of the freeze seal and freeze seal equipment used to isolate the containment spray system leakage on July 23, 1996.

b.

Observations and Findin s The freeze seal equipment used on the containment spray piping weighed approximately 15 lbs. and was installed on a horizontal section of the 2 inch test line approximately 6 feet from the leaking pipe joint. On one side of the freeze seal, the CS pipe is firmlysupported by a concrete shield wall that the pipe penetrates, and the other side is welded into the 10 inch CS and Sl pump suction piping.

The licensee issued design analysis DA-ME-96-069, Rev.0, "Containment Spray Test Line Leak," on July 23, 1996.

The analysis described the seismic response of the CS test line with the freeze seal equipment installed, and determined that the combined maximum pipe stresses under safe shutdown earthquake conditions would be approximately 47% of the code allowable stresses for this pipe.

The analysis also evaluated the ability of the freeze plug to maintain a pressure boundary under seismic conditions.

Those results indicated that the combined pressure and piping forces imposed by seismic conditions were approximately 8.6% of the holding power of the freeze plug.

Based on these results, the analysis concluded that the freeze seal equipment and the freeze plug maintained the seismic qualification of the CS test line, and provided a acceptable pressure boundary.

The licensee also issued design analysis DA-ME-96-071, Rev.0, "Temperature Profile of Containment Spray Test Line for Freeze Plug," on July 23, 1996.

This analysis was performed to evaluate the thermal conditions in the CS test with the freeze seal installed, and under maximum temperatures following a design basis loss of coolant accident (LOCA). With an ambient air temperature of 120 F and CS system temperature of 286 F, the analysis showed that the piping at the freeze seal was only slightly higher than ambient air temperature.

The licensee also issued a 10 CFR 50.59 Safety Review of the freeze seal installation and concluded that all effected plant equipment remained within analyzed design parameters based upon the engineering evaluation C.

Conclusions The inspector concluded that both design analyses employed conservative engineering assumptions to evaluate the acceptability of the freeze seal on the CS test line. The analyzed pipe stresses were well within the maximum code allowables, and the seal was capable of maintaining a good pressure boundary under both seismic and post-LOCA conditions.

The Safety Review contained sufficient justification for concluding that the freeze seal application was within previously analyzed conditions.

Anal sis of Main Feedwater Valve V-3984 Tem orar Leak Re air a0 Ins ection Sco e 62703 The inspectors reviewed the engineering evaluations associated with the licensee's plan to perform temporary on-line leak repair of the Main Feedwater (MFW) block valve V-3984 as discussed in sections 02.4 and M1.1.2.

b.

Observation and Findin s Valve V-3984 is a 14 inch manual gate valve on the inlet side of the B-MFW regulating valve (AOV-4270) in the turbine building. V-3984 does not perform a safety-related function, and is located in the non-safety related portion of the MFW System.

However, the valve is located in a portion of the system that is classified as seismic category 1A. Due to normal operating conditions of approximately 1000 psig and 425 degrees F at full power operation, V-3984 is in a section of MFW piping that is classified as high energy.

Since V-3984 is a manual gate valve its only safety-significant function is to maintain pressure boundary integrity of the MFW piping.

On August 13, 1996, the valve leakage became significantly worse in a short period of time and caused steam to issue from all 12 holes in the valve body.

After the reactor tripped on August 20, 1996, the licensee hired a contractor to assess the valve condition and propose an on-line valve leak repair method.

The contractor proposed two methods to repair the leaking valve.

The first method was to attempt to seal the leak on-line through twelve existing knock-out holes present in the valve body using an approved leak sealant.

The contractor proceeded with the first repair method and tapped out the existing retaining ring knock-out holes to allow for the installation of leak repair injection valves.

The licensee's analysis showed that these tapped holes would have no affect on the designed pressure rating of the valve since these holes were contained in the original valve design.

The existing holes would allow for the application of a phenol resin, an approved-for-use leak repair material.

This material is applied under pressure through the installed injection valves into the pressure seal ring area of the valve which would serve to stop any further leakage from the valve.

The leak repair method was partially successful in that it slowed down the leak on the valve perimeter.

However, it diverted the leak path to the upper bonnet area around the

packing gland and the bonnet bolts, and still resulted in substantial water and steam leakage from the top of the valve.

The proposed second method of sealing the valve would also be on-line and would involve performing the following actions:

I

~

Drilling four 3/16" diameter holes through the valve in the pressure boundary region of the valve body at the bottom tip of the valve pressure seal ring, and

~

Injecting a thermosetting sealant into the valve body through the drilled holes.

The licensee performed an evaluation to address engineering and operational issues associated with a planned on-line leak repair of the MFW manual isolation valve V-3984.

After the first seal repair was not successful in stopping the leak, the NRC became concerned about the second repair method since it appeared that RGRE engineering had not fully evaluated the potential for steam cutting of the bonnet bolts, and had not performed stress calculations of the seal injection pressures in the vicinity of the valve seal ring. The plan to inject a sealant through additional holes to be drilled in the pressure retaining portion of the valve body could cause additional stress on the bolts and potentially degrade the integrity of the pressure boundary.

The licensee's analysis concluded that the second method of on-line leak sealing would not affect the pressure boundary integrity of V-3984 and would not have an impact on the operation of the MFW System because the sealant would be limited to the local area of the bonnet adjacent to the drilled holes, Also, the additional weight of the sealant would not impact the seismic qualification of the valve.

Engineering determined from manufacturing design data that the bonnet and valve body bolting provided for V-3984 does not perform a pressure retaining function by design.

The pressure retaining function was provided by the pressure seal ring feature inside the valve.

The bolts (type A193 B7) were used primarily for initial preloading of the pressure seal ring during valve assembly.

The bolt material is a high strength alloy steal but is only susceptible to corrosion attack from borated water.

Since there is no borated water in the MFW System this issue is not a concern.

Since the bolts were resistant to steam cutting and the portions exposed to the steam leak were in a pool that quenched the steam leak, the licensee's analysis concluded that the bolts were not susceptible to steam cutting.

However, after several conferences with the valve manufacturer and its contractor the licensee determined that the second method (drilling holes in the valve body)

may not be effective, and that the leak may not be correctable by any means other than valve disassembl C.

Conclusion Once the valve leakage increased significantly, the licensee made adequate preparations for an on-line repair. After the first repair attempt was not successful, the licensee's decision to monitor the leakage and defer the repairs to an outage was prudent.

E7 Quality Assurance in Engineering Activities E7.1 Review of the Licensee's 10 CFR 50.59 Pro ram a.

Ins ection Sco e (37001)

The NRR Project Manager for the Ginna Station visited the RGRE corporate offices and the plant site on July 23 - 25, 1996, to review the licensee's 10 CFR 50.59 process, and to attend plant operations review committee (PORC) meetings where 50.59 evaluations were presented and discussed.

The review of the licensee's 10 CFR 50.59 process followed the guidelines set forth by NRC Inspection Manual Inspection Procedure 37001, 10 CFR 50.59 Safety Evaluation Program.

The review included the review of the procedures for conducting safety reviews and safety evaluations; review of the training program; review of associated procedures; and a review of samples of Plant Change Records (PCRs), Procedure Change Notices (PCRs) and Safety Evaluations.

b.

Observations and Findin s

~ Procedures for Conducting Safety Reviews and Safety Evaluations Nuclear Directive ND-SEV, "Safety Evaluations," establishes the requirements for Safety Evaluations, operability determinations, and Notices of Enforcement Discretion, and assigns responsibility for implementing those requirements.

The directive requires all proposed changes receive a safety review. This is a screening process to determine if a written 10 CFR 50.59 Safety Evaluation is required.

Individuals performing the review and approval of Safety Reviews and 10 CFR 50.59 Safety Evaluations are required to be trained and qualified to perform such activities.

A Safety Review in accordance with IP-SEV-1, "Preparation, Review and Approval of Safety Reviews," is performed for activities that are considered to constitute changes to the facility. These typically result from plant modifications, proposed plant procedures, plant procedure changes, disposition of nonconformance reports, changes to the plant design basis or design basis documents, change to controlled configuration drawings or documents, changes to the Updated Final Safety Analysis Report (UFSAR), changes to a components's safety classification, and changes to or deletion of a regulatory commitmen The Safety Review determines the need to prepare a written 10 CFR 50,59 Safety Evaluation.

It also determines the need for PORC review of the change.

The procedure directs six questions that require a "yes" or "no" answer, and each requires a basis explanation.

A yes answer to any question requires the preparation of a safety evaluation.

It was noted that question 4, "Willthe change or condition alter the design, function or method of performing a function of a structure, system or component as described in the Safety Analysis Report or in an NRC SER?"

depended upon an attachment for guidance for answering the question.

The attachment addressed changes related to procedures, but did not address changes related to plant modifications.

However, this discrepancy was also recognized by the Nuclear Safety and Licensing Department which was implementing a change to the procedure to correct the issue.

The procedure also required the review and approval of a qualified reviewer who would make the determination whether the change requires PORC review. This determination is based on the judgment of the qualified reviewer and there does not appear to be any written guidance for making the determination.

The procedure is relatively simple and the completion of the procedure is highly dependent upon the training received by the preparer and the reviewer.

Interface Procedure IP-SEV-2, "Preparation, Review, and Approval of 10 CFR 50.59 Safety Evaluations," provided instructions and a format for the preparation, review, and approval of 10 CFR 50.59 Safety Evaluations for determining whether or not proposed changes (1) involve an unreviewed safety question or change to the Technical Specifications (TS), and (2) may be implemented without NRC approval.

The procedure ensured that proposed plant changes were evaluated with respect to original safety considerations of the plant design basis while preserving the Safety Analysis Report (SAR) and the TS. The procedure provided the necessary information used in responding to the 3 conditions that define an unreviewed safety question.

Completion of the procedure was highly dependent on the details of the change including plans and analyses, on the training of the preparer and the qualified reviewer and on the references that provide additional guidance.

The inspector noted that NSAC-125, "Guidelines for 10 CFR 50.59 Safety Evaluations," is a primary guidance document referenced by IP-SEV-2.

The procedure leads the preparer and reviewer to a determination of whether the change should involve a change in the UFSAR. The procedure explicitly identified the documents that comprise the UFSAR in the context of the 10 CFR 50.59 rule. The list not only identified the UFSAR, the TS and its bases, the operating license and license conditions, but also NRC Safety Evaluations and correspondence referenced therein.

RGKE correspondence submitted to the NRC in support of approved license amendments and RGRE correspondence used as referenced or source documents in the UFSAR were also referenced on the list.

Each Safety Evaluation required review and approval of a qualified reviewer, and the manager of Nuclear Safety and Licensing (NSKL). The inspector noted that NSRL assists in the preparation of Safety Evaluations and played a major roll in the preparation of the procedures and training for the preparation of Safety Reviews and

Safety Evaluations.

In general the procedure for making 10 CFR 50.59 determinations was found to be very good.

~ Training for Preparation of Safety Reviews and Safety Evaluations The licensee has a formal training program for the preparation and review of Safety Reviews and Safety Evaluations.

It consists of 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> each of class room instruction for Safety Reviews and Safety Evaluations.

Based on personnel interviews which included discussions on their understanding of the 10 CFR 50.59 process and implementing procedures, the inspector determined the training was very good.

Much of the training is by example.

Nuclear Safety and Licensing developed the lesson plans and conducts the training.

Since NSRL was key in the development of the procedures for Safety Reviews and Safety Evaluations and in the review of Safety Evaluations, NSRL provided assistance in the preparation of Safety Reviews and Safety Evaluations.

The training is oriented around the NSAC-125 guidelines.

The licensee has also prepared a Self-Study Training Guide for those not able to attend the formal class room program.

Since the procedures for performing Safety Reviews and Safety Evaluations are relatively new, the licensee put forth a concerted effort in March 1995 for the training of personnel

~ To date, there are 120 people that received training in Safety Reviews and Safety Evaluations and are qualified reviewers.

The licensee does not have a refresher training program in place but indicated that it was under consideration.

~ Other Associated Procedures Other procedures that relate to the 10 CFR 59.59 process were reviewed and included the following:

IP-DES-2, Plant Change Process IP-LPC-1, Commitment and Action Tracking IP-LPC-2, Updated Final Safety Analysis Report and Associated Documents Control EP-3-S-306, Change Impact Evaluation Form IP-DES-2 is the procedure to provide overall guidance for the plant change process from conceptual design through close-out of the change.

The procedure covers the processing of a Plant Change Record (PCR) which is used to document plant changes.

The assigned engineer performs a Safety Review of the proposed plant change and this review is retained with the PCR package.

When required by the Change Impact Evaluation or the Safety Review, the assigned engineer will prepare a Safety Evaluation.

The Safety Evaluation is not necessarily a part of the PCR package.

However, the PCR package is most useful in the preparation and review of the Safety Evaluation.

The Change Impact Evaluation also determines if the change requires an UFSAR change.

IP-LPC-2 establishes the instructions for processing changes and revisions to the UFSA ~ Review of 10 CFR 50.59 Determinations The annual report for August 1994 through July 1995 was reviewed.

It identified 34 PCRs, 1 procedure change and 1 test.

The annual report indicated that all modifications, procedure changes and tests did not constitute an unreviewed safety question per the criteria of 10 CFR 50.59.

Since the contents of the annual report consisted of many changes related to the steam generator project, and since these have been separately reviewed by the staff during the steam generator replacement project, changes initiated since July 1995 were selected for review.

In addition, such changes would likely involve Safety Reviews and Safety Evaluations performed under the requirements of current procedures.

The following PCRs, procedure changes, and Safety Evaluations were reviewed:

PCRs95-028 95-039 96-065 96-082 Installation of access doors on splash shield for the RHR pump motors and motor generator sets Replace the Satellite Station A (SSA) and Satellite Station C (SSC)

Batteries.

Modify actuator non-locking gear set to locking gear set Containment spray piping leak Procedure Chan e Notices PCNs M-15.1.3 0-1A RE-20.1 CH-PRI-CCW-LEAK Determination of CCW System Leakage FPS-5 Inspection, Lubrication and/or Spring Testing of PIV and Curb Valves Inspection, Testing and Setup of Diesel Generator Governors Filling and Venting the RCS Precision Flow Calorimetric Safet Evaluations SEV-1052 SEV-1058 SEV-1063 SEV-1029 SEV-1057 RWST Modification Addition of Automatic Turbine Trip on High Vibration Use of Containment Equipment Hatch Enclosure Building (PCR 96-019)

Instrument Set Point Changes Associated with Steam Generator Replacement and Reduced RCS average Temperature 18 Month Fuel Cycle The inspector reviewed the Safety Reviews for the above PCRs and Procedure Changes.

All followed the standard format specified in IP-SEV-1, and answered the 6 questions in the procedure and determined that a Safety Evaluation and review by PORC was not necessary.

Except for PCR 95-028, "Access Doors for RHR Motor Covers," each determination appeared to be properly evaluated.

Question 4a, "Will the change or condition alter the design, function or method of performing a function of a structure, system, component or procedure as described in the SAF or

in an NRC SER?" was answered

"no" for PCR 95-028.

However, the bases for providing that answer was not thoroughly explained.

Although engineering judgment would likely determine that this did not involve an unreviewed safety question, the answer was not entirely proper and needed to be checked in the "yes" block or the bases expanded to explain why it was not a change in the design.

The inspector also reviewed the Safety Evaluations identified above and, with one exception, noted the Safety Evaluations appeared to stand alone without the need of supporting documentation.

For SEV-1052, "RWST Modification," it was necessary to review the PCR package to obtain a full understanding of the modification configuration.

c.

Conclusions The licensee has established a good 10 CFR 50.59 determination process.

The adequate implementation of the process is highly dependent on the training of people preparing and reviewing the Safety Reviews and preparing and reviewing the Safety Evaluations.

The training program was determined to be very good.

The training is accomplished by a structured training program in which many engineers have been trained and many have been certified as qualified reviewers.

R1 Radiological Protection and Chemistry (RP8cC) Controls R1.1 Radiation Monitors Near the New Fuel Stora e Area a.

Ins ection Sco e (83750)

The inspectors reviewed the licensee's conformance with 10 CFR 70.24, "Criticality accident requirements."

b.

Observations and Findin s 10 CFR 70.24 requires that licensees authorized to possess uranium-235 in quantities exceeding 700 grams have a monitoring system that is capable of detecting a criticality in all areas where this material is stored.

The monitoring system must be capable of detecting a criticality that produces an absorbed dose in soft tissue of 20 rads of combined neutron and gamma radiation at an unshielded distance of 2 meters from the reacting material within one minute.

The rule also requires that for licenses issued prior to December 6, 1974, a monitoring system capable of detecting a criticality that generates radiation levels of 300 rems per hour one foot from the source of the radiation.

An additional requirement stipulated that the monitoring system shall have a preset alarm setpoint of not less than 5 mR per hour but not more than 20 mR per hou The radiation area monitor (R-5) in the auxiliary building at the Ginna Station is located on a wall at the northeast corner of the spent fuel pool (SFP).

The new fuel preparation area is approximately 20 to 25 feet from this detector.

The detector is currently set to alarm at 25 mR per hour, and is placed so that a high radiation level near the SFP or the new fuel preparation area could be readily detected.

The inspector discussed the criticality monitoring capability of R;5 for the new fuel preparation area with the licensee's senior health physicist and the reactor core system engineer.

Preliminary evaluations were not conclusive as to whether the R-5 monitor is currently.set up to indicate at the levels specified by the rule. The instrument is capable of detecting a radiation field of 10'R/hr and may have to be reset to the provide the criticality monitoring required by the rule.

The licensee's review was ongoing at the end of this inspection.

10 CFR 70.24 contains a provision for an exemption to the rule in cases where good cause exists.

The licensee is confident that the R-5 monitor would alarm for any inadvertent

'riticality near the SFP or new fuel preparation area.

However, in'the event. that the R-5 monitor cannot satisfy the specific requirements of the rule, the licensee will request an exemption.

The next refueling outage at the Ginna Station is presently

= scheduled for late September 1997, and new fuel will not,be stored in the auxiliary building before mid July 1997.

c.

Conclusions The inspectors concluded that the licensee's efforts to evaluate the capabilities of the R-5 detector were appropriate.

The licensee will either verify the R-5 detector meets the requirements, or request an exemption from the rule, prior to the next time new fuel is stored on site.

F1 Control of Fire Protection Activities F1.1 Housekee in in the Auxilia Buildin a.

Ins ection Sco e (62703)

The inspectors conducted a routine housekeeping tour of all levels of the Auxiliary Building.

b.

Observations and Findin s

'i During an inspection tour of the auxiliary building on August 13, 1996, the inspectors identified a notable buildup of dust on and around the spent fuel pool, and on the screens covering the air vents of energized transformers inside safeguards buses 14 and 16. The spent fuel pool skimmer and the auxiliary building air handling unit were out of service and not operating at the time. Air flow to the safeguards transformers was not completely blocked by the dust buildup; however, it could have caused insufficient ventilation if allowed to persist or increase.

These

'a

conditions were pointed out to the licensee, who took immediate action to clean around the spent fuel pool and the air vent screens.

Maintenance procedure M-1306, "Ginna Station Material Condition Inspection Program," included a general housekeeping task for inspection of electrical equipment in the auxiliary building and had been scheduled for the following week, but it was rescheduled for the following day to address the inspectors'oncerns.

The procedure included a specific inspection point for dust, dirt, and debris on electrical equipment and panels, but relied on the skill of the electrical maintenance craft to determine what constituted a discrepancy.

The air vents were cleaned and no indications of overheated equipment were evident on buses 14 and 16.

The electrical maintenance foreman indicated that all air vents for safeguards transformers were cleaned during the past 1996 refueling outage, and that activities in the auxiliary building toward the end of the outage may have generated high levels of dust that became trapped in these vents.

Procedure M-1306 was performed weekly in one of 12 plant zones, but it had not yet been performed in the auxiliary building.

The foreman stated that the next time M-1306 is performed, an assessment will be. made to determine if the frequency of performing the procedure is appropriate for electrical equipment in the auxiliary building.

He noted that non-safeguards buses 13 and 15 currently collect enough dust to be inspected and cleaned on a weekly basis, and that buses 14 and 16 should be evaluated to see if more frequent inspection is warranted.

c.

Conclusions The inspectors concluded that the licensee took prompt action to clean the areas noted in the auxiliary building, and to assure that no equipment damage resulted from a lack of adequate ventilation.

The maintenance procedure for inspecting the material condition of electrical equipment was adequate; however, a review of it'

frequency for the safeguards buses was appropriate.

V. Mana ement Meetln s

X1 Exit Meeting Summary An exit meeting was held on July 26, 1996 with the plant management for the 10 CFR 50.59 program review.

An exit meeting for the remainder of the inspection activity was held on September 17, 1996.

X3 Management Meeting Summary X3.1 Senior NRC Mana ement Visits to The Ginna Station On the dates indicated, the following senior NRC managers toured the station and interviewed various personnel:

~ August 13 5 14, 1996:

H. Miller, Regional Administrator, Region I

~ August 20 5 21, 1996:

A. Blough, Asst. Director, Division of Reactor Safety, Region I

J: Mitchell, Acting Director, Project Directorate l-1, NRR B. Sheron, Director, Division of Engineering, NRR