IR 05000352/1997003

Insp Repts 50-352/97-03 & 50-353/97-03 on 970330-0527. Violations Noted.Major Areas Inspected:Operations,Maint, Engineering & Plant Support

ML20149E146
Person / Time
Site:	Limerick
Issue date:	07/11/1997
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20149E126	List: IR 05000352/1997003 ML20149E121 ML20149E139
References
50-352-97-03, 50-352-97-3, 50-353-97-03, 50-353-97-3, NUDOCS 9707180121
Download: ML20149E146 (28)

Text

. -. ... - - -- .- . - . . . - - . - . . . , . . - . . - - . - . - - .. - .

.-

!

,

..

U.S. NUCLEAR REGULATORY COMMISSION

'

j REGION I

!

!

'

!

L

!

'

l

'

License Nos. NPF-39 i NPF-85 l

!

l l Report Nos. 97-03 97-03 Docket Nos. 50-352 50-353 4 i

a Licensee: PECO Ener0y Correspondence Control Desk P.O. Box 195 l

. Wayne, PA 19087-0195

!

'

.

Facility Name: Limerick Generating Station, Units 1 and 2 Inspection Period: March 30,1997 through May 27,1997 i i

'

inspectors: N. S. Perry, Senior Resident inspector F. P. Bonnett, Resident inspector -

G. C. Smith, Senior Security Specialist Approved by: Richard R. Keimig, Chief l Projects Branch 4

! i

!

,

l 9707190121 973711 PDR ADOCK G50fC352 G pg 1:6

. ,. , ,_

. - .- . - - .- .-. .-- - .- - . .- - . - - - . - - - - - .

!

,

..

l EXECUTIVE SUMMARY l Limerick Generating Station, Units 1 & 2

NRC Inspection Report 50-352/97-03,50-353/97-03 l This integrated inspection included aspects of PECO Energy operations, maintenance, i engineering, and plant support. The report covers an 8-week period of resident inspection; l 'in addition, it includes the results of an announced inspection by a regional senior security specialist.

i Operations -

o Conduct of operations was professional and safety-conscious (Section 01.1).

i e A reactor operator (RO) mispositioned the Unit 1 Load Limit Potentiometer during an evolution to return the turbine-generator to service. A second RO inadequately i verified the potentiometer to be in the correct position during a subsequent

"

surveillance test which resulted in a reactivity event at Unit 1. A root cause for the event was inadequate operator understanding of the function of the Load Limit

,

Potentiometer and its impact on reactor reactivity (Section 01.2).

l

e On September 23,1995, plant management did not shut down Unit 1, as specified

] by proceduro GP-12, when the chemistry Action Level 3 for sulfates was exceeded.

. Although continued operation at low power with the high sulfate levels did result in j some small degradation of vessel components, this was the accepted method for returning chemistry levels to acceptable values. The procedure was temporarily j changed in a timely manner to remove the shutdown requirement. However, this i- process was inappropriate since the change involved an intent change the j- procedure. Technical Specification (TS) 6.8.3 prohibits temporary changes for j changes to the intent of procedures in order to assure proper safety reviews are

performed prior to implementation. For this instance, the safety consequences were

! low. The improper use of the temporary change constitutes a violation of TS 6.8.3 ;

(Section 3.1). )

! e Operator performance in restoring the RWCU system to service following the i. system outage was excellent. Procedures, although adequate, were not specific to i 1 the evolution and needed to be temporarily changed, and were not well human  !

) factored in that they contained more than one action per step (Section.04.1). ;

i

'

!

Maintenance

'

e The two turbine outages to repair the Unit 2 main generator flexib;e coupling and Alterex bearing were well planned and implemented. The turbine vibration data following the outages indicated a substantial decrease in vibration levels (Section j M 1.1 ).

!

e The inspectors concluded that the Limerick staff performed the 18 month inspection !

and maintenance of the D22 EDG well(Section M1.2).

ii

,_ ,_ -

.-

.

Executive Summary

• The Unit 1 safeguards battery replacement activities were performed safely and within the requirements of the Technical Specifications (Section M1.3).

• Observed surveillance tests were conducted well using approved procedures, and were completed with satisfactory results. Communications between the various work and support groups were good, and supervisor oversight was good (Section l M1.4). ,

I

• The inspector concluded that the late performance of a service water valve circuit l contact test was weak, but there were no safety consequences associated with it, I since the valve was always operable after restoration. Two other surveillances i

'

reviewed exhibited delays in resolution due to poor communications and tracking.

Corrective actions are being evaluated (Section M1.5).

• I&C technicians demonstrated an excellent questioning attitude which led to l identifying a discrepancy in the calibration parameters for the hydrogen recombiners between the two units (Section M3.1).

e Operations management enhanced the method of performing the Reactor Enclosure Secondary Containment Integrity Verification, after evaluating several concerns raised by the inspector. The issue will be discussed with personnel from the Peach Bottom Station, and will remain unresolved as to the test's conformance with Technical Specifications (TSs) until after the discussion (50-352,353/97-03-01)

(Section M3.2).

Enaineerina

• The inspector concluded that the or;ginal engineering analysis for six untested steam dampers was inadequate, since four failed; however, it was reasonable, since no dried grease was identified on any of the other damper linkages in February 1997, and therefore failure could not have been reasonably expected. The failure of the four dampers is another example of the non-cited violation described in NRC Integrated Inspection Report 50-352/97-01, 50-353/97-01; Notice of Violation (Section E2.1).

Plant Suonort

• Suppression pool water was not directly injected into the reactor in September 1995. Suppression pool water was pumped to the hotwell, and then to the reactor, but this was acceptable and the water was probably cleaned first (Section R1.1).

• Log entries indicate that reactor water let down to the CST in January 1993, was appropriate and adequate precautions were taken, and the appropriate personnel were involved. Also, this activity has been properly performed a number of times, without problems. Activity levels were below the administrative limits for the CST, and the CRD filters were affected by the water, but this is a radiological concern, t

j iii

..

..

Executive Summary and appropriate precautions are taken during replacement activities. Chemistry personnel searched to see if the administrative limit for the CST activity was exceeded at another time, but did not identify any instances (Section R1.1).

• The Chemistry Manager and the BOP Chemistry Manager did not meet the minimum requirements for nuclear experience for some period of time after appointment to their positions in 1993; however, other chemistry personnel met the requirements ;

and currently all chen.istry management meet the requirements. In July 1995, a I procedure was implemented (A-C-031, Nuclear Staff Qualification Requirements and Organizational Structure Control), which provides administrative controls for controlling and maintaining the qualification level for the Nuclear Generation Group.

The failure to meet the minimurn experience requirements for two chemistry management personnel was a violation of Technical Specification 6.3.1. This failure !

is being treated as a non-cited violation (Section R5.1).

• Unauthorized keys to locked high radiation areas, not under the administrative control of Shift Supervision on duty and/or health physics supervision, were l available to personnel (fire protection personnel) since about 1993 through about '

November 1996. This resulted in a violation of Technical Specification 6.12.2, which requires in part, that areas accessible to personnel with radiation levels such that a major portion of the body could receive in 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> a dose greater than 1000 mrems shall be provided with locked doors to prevent unauthorized entry, and the keys shall be maintained under the eininistrative control of Shift Supervision on duty and/or the health physics tupervision. Review of the investigation into the improper control of vital area and locked high radiation area keys determined that the investigation was thorough, complete and appropriate, and timely compensatory measures were irnplemented until the potentially compromised locks could be changed out. Based on the fact that the individuals that had the unauthorized grand rnaster keys did not know they would open vital area or high radiation doors and because the vital area doors are alarmed and monitored, the unauthorized grand master keys were determined not to have posed an unmonitored security vulaerability (Section S8).

i l

l

,

iv

.

..

..

Report Details Summarv of Plant Status Unit 1 began this inspection period operating at 100 percent power. The unit remained at full power throughout the inspection period with minor exceptions for testing and the following operational event:

• April 11 Operators reduced reactor power to 25 percent to remove the main generator from the grid and repair a hydrogen leak at the main generator's C phase neutral high voltage bushing. Operators maintained reactor power at 25 percent using the turbine bypass valves until the generator was synchronized to the grid on April 15.

Operators raised and held power at 60 percent to perform condenser waterbox maintenance. After adjusting the control rod pattern, operators increased reactor power to 100 percent using a ramp rate of 3 percent per hour. The unit achieved full power operation on April 18.

Unit 2 operated at 70 percent power at the beginning of this inspection period due to high vibration and shaft ali]nment problems identified at the No.11 bearing of the turbine-generator. The unit was operating at 100 percent power at the end of the period. The following describes the operational events that occurred during this period:

• March 30 Operators reduced reactor power to 22 percent to remove the main generator from the grid to inspect and repair the Alterex flexible coupling on the main generator. Operators maintained reactor power at 22 percent using the turbine bypass valves until the generator was synchronized to the grid on April 4. Engineering installed extra monitoring equipment on the bearing and A!terex housing to observe and collect performance data. Operators returned the unit to full power on April 6.

• April 16 Operators reduced reactor power to 45 percent due to in:reasing vibration indicated at the No.11 Alterex bearing. Operators observed the bearing at 8 mils and increasing, and entered the off-normal procedure for high vibration. Bearing vibration reached as high as 10 I mils prior to reducing reactor power. Management formed a task force of engineers and technicians to evaluate the impact of the 1 vibration on continued turbine-generator operation. The task force could not identify a problem; however, additional monitoring i equipment was installed to monitor generator vibration and ;

performance during a slow deliberate power ascension. Operators slowly restored power to 100 percent on April 19. )

I

• April 24 Operators reduced reactor power to 70 percent as a precaution due to ]

'

increasing vibration at the No.11 Alterex bearing. The engineering task force performed an in-depth review to determine the root causes and to develop a plan to correct the problem (which was to remove ,

the generator from service on April 30). After monitoring and

).

l

- . . . .. , -. -- - . - . ~ = - - -. ..

.'

i ..

assessing key generator parameters, the task force was confident that power could be raised to 85 percent until the planned outage. The

operators increased reactor power to 85 percent on April 28.

l

! * April 30 Operators reduced reactor power to 23 percent to remove the main generator from the grid, disassemble the Alterex, and replace the No.

11 and 12 Alterex bearings on the main generator. Operators maintained reactor power at 23 percent using the turbine bypass l valves until the generator was synchronized to the grid on May 6.

Operators raised reactor power to 100 percent on May 7.

l. Operations 01 Conduct of Operations' -

l 01.1 General Comments (717071

'

Using Inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing plant operations. In general, PECO Energy's conduct of operations was professional and safety-conscious. Operators conducted routine activities well. Operators demonstrated good communication skills and control during plant evolutions, including the restoration of the reactor water clean-up (RWCU) system at Unit 1, the power reduction to 25 percent at Unit 1 for the generator bushing repair, and several power reductions to 25 percent at Unit 2 for generator bearing repairs.

01.2 Unexoected Power Dron Durina Turbine Control Valve Testina - Unit 1 a. Insoection Scoce (71707)

During the performance of a Unit 1 turbine control valve (TCV) functional test on April 26, a larger than expected decrease in reactor power and unexpected increase in reactor level occurred. The unit was operating at 92 percent power as specified by the surveillance test when the event occurred. The inspector reviewed and discussed the event with the system manager and several control room operators, b. Observations and Findinas The event occurred while performing the exercise test for the No.1 TCV of surveillance test (ST)-6-001-765-1, Main Turbine Control Valve Exercise and Reactor Protection System (RPS) Channel Functional Test. The TCV was reopening when the Reactor Operator (RO)

noted that the standby electro-hydraulic control (EHC) pump had started, a high vibration spike occurred at the No.1 turbine bearing, reactor power dipped to about 75 percent, and l

l

! ' Topical headings such as 01, M8, etc., are used in accordance with the NRC standardized i reactor inspection report outline. Individual reports are not expected to address all outline topics.

.-

..

reactor level swelled to about 39 inches. The plant stabilized after the No.1 TCV reached the full opan position. The RO immediately notified the control room supervisor (CRS),

stopped further testing, ard called the system manager to investigate the cause of the unexpected plant response.

The system manager determined that the Load Limit Potentiometer, on the EHC control panel, was mispositioned and was the cause of the event. The potentiometer is normally set for 8.44 to limit the operation of the No. 4 TCV to 70 percent open, but the system manager found it set at 9.44. The system manager also determined that the potentiometer was most likely set incorrectly during the re-start of the turbine following the neutral bushing repair activity on April 15. The potentiometer was reset correctly and the test was performed without further problem.

The Performance Enhancement Program (PEP) review initiated for the event, concluded that the root causes of the event were due to the RO not being adequately trained on the use of the potentiometer (mis-setting the limit), and that the procedures did not address the potentiometer's method of operation. A training bulletin to all operations personnel highlighted the above concerns. The inspector, however, did nut totally agree with the PEP's conclusions and training bulletin's focus.

The inspector found that a sample of operators did not understand the basis of the 8.44 setpoint, and of the potential impact a mispositioned Load Limit Potentiometer had on plant operations. The main turbine operates in a partial-arc steam admission mode, which opens the Nos.1 - 3 TCVs together to approximately 89 percent before opening the No. 4 TCV.

The Load Limit Potentiometer limits the TCVs from admitting steam beyond 105 percent load and ensures that the turbine bypass valves function to reduce the affects of the

" velocity limiting" of the TCVs. W;th the potentiometer set at a higher value, the upper TCV operating limit is raised allowing the TCV to open further, which willincrease the magnitude of pressure oscillation effects on reactor power.

For this event, after the RO stroked closed the No.1 TCV, the No. 4 TCV had opened further. The TCV strokes open at a faster rate (in about 10 seconds) than closed (about 35-40 seconds). As the No.1 TCV stroked open, a larger inlet for steam admission to the turbine existed resulting in the lowering of reactor pressure, causing tik. power drop; the power drop initiated the level swell; the excess steam admission to the turbine also caused the high bearing vibration spike; and the extra control oil demand to shut the No. 4 TCV while the No.1 TCV was opening started the standby EHC pump. The inspector determined that the plant response was normal for this scenario.

The inspector concluded that the procedures were incomplete, in that they did not give the operator adequate guidance ao to the potential transient and plant impact on reactivity if mis-operation of the potentiometer occurred. Further, operator training was lacking concerning the function of the Load Limit Potentiometer and its importance to limit TCV operation in a partial-arc admission turbine.

Operations management continued to implement corrective actions regarding this issue.

i Training will discuse the event and its causes with all operators during upcoming requalification training sessions. The ST procedures are being further reviewed and revised

.

..

to incorporate cautions and notes regarding the potentiometer and its plant impact. The inspector noted that the Operations Department procedure review identified an inconsistency between the ST for Unit 1 and Unit 2. The potentiometer setpoint was specified in the Unit 1 ST, but was not in the Unit 2 ST. This inconsistency was corrected, c. Conclusions An RO mispositioned the Unit 1 Load Limit Potentiometer during an evolution to return the turbine-generator to service. A second RO inadequately verified the potentiometer to be in the correct position during a subsequent surveillance test causing a reactivity event at Unit 1. A root cause for the event was the inadequate operator understanding of the function of the Load Limit Potentiometer and its impact on reactor reactivity.

03 Operations Procedures and Documentation 03.1 Procedure Chanae to GP-12 a. Inspection Scope (71707. 71750)

The inspector reviewed the circumstances surrounding an instance where General Plant ,

procedure GP-12, Water Chemistry Control Action Plan, was changed using the temporary change process to apparently avoid a required plant shutdown during the startup of Unit 1, on September 23,1995. The inspector reviewed GP-12's requirement with respect to chemistry control and plant shutdowns for that date, and the Chemistry and operator logs for indications of compliance. The inspector also reviewed a change to GP-12 on September 23,1995, and its timing with respect to operator knowledge of adverse chemistry findings.

b. Observations and Findinos On September 23,1995, Unit 1 was starting up, after being shut down for safety relief valve work and suppression pool cleaning. Shortly after the reactor was adding heat to the reactor coolant system, chemistry personnel noted an increasing trend on conductivity and sulfates. Chemistry logs indicate that Action Level 3 of GP-12 was entered at 0525, based on a reactor water sulfate level of 207 ppb (Action Level 3 is when sulfate is greater than 200 ppb). Operation logs for September 23,1995, indicate that Action Level 3 was entered at 0641.

On September 22,1995, GP-12, Water Chemistry Control Action Plan, stated, for Action Level 3, to immediately begin a controlled plant shutdown using GP-3. Additionally, it stated to be in at least Hot Shutdown within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, and in Cold Shutdown within the next 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> regardless of whether the parameter drops below the Action Level 3 value during the shutdown. There were no log entries for starting a controlled shutdown or entry into GP-3. Operation togs imply that power was maintained at approximately 2-3 % until 1610 on September 23,1995, when power was increased. Currently, management expectations are that a required plant shutdown would be logged properly, especially if it is the result of exceeding a technical specification LCO.

. . _ _ _ m. _ . _ . . _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ .

.. ,

..

The Shift Manager on September 23,1995, recalls that they did not shut down the unit, but that they did contact reactor engineering personnel to get direction on how best to shut down the plant. He indicated that he believes they may have entered GP-3, without logging the entry, but they never reduced power, because the GP-12 procedure was temporarily changed first. In GP-3, power reduction does not take place until page 9 of the procedure, after Precautions and other steps are performed. The Shift Manager remembers that GP-12 was changed early in the shift, which began at 0630. There is no record of the ,

time the procedure was changed, only the date.

The inspector's review raised several concerns with the temporary change to GP-12, which was completed on September 23,1995. First, the step directing an immediate plant shutdown was changed to direct consultation with chemistry management and the plant Vice President to evaluate and determine if a plant shutdown is required. This is an intent change to the procedure, requiring a procedure revision, with a 10 CFR 50.59 review and a PORC review prior to implementation, rather than a temporary change eventually becoming '

a permanent change. Additionally, the same person signed the temporary change as the

- originator and the Station Qualified Reviewer, and the Responsible Superintendent was not the right person; similar issues to these were identified by the inspectors and documented in NRC Integrated Inspection Report 50-352/96-07, 50-353/96-07; which resulted in a violation. This programmatic problem has been addressed by plant management. Since -

1995,'the administrative procedure form for temporary changes was revised to include specific questions to determine if a procedure change is an intent change. Additionally, plant management has recently reinforced, at various plant meetings, the need to pay close attention to intent change determinations.

I For this event, safety consequences were low. PECO concluded that operation with the elevated sulfate levels resulted in an impact on reactor life equal to one year of crack growth under normal water chemistry conditions, due to intergranular stress corrosion cracking. Through consultations with General Electric and the Electric Power Research Institute, PECO confirmed that the best method of eliminating the sulfate was to decompose the resin with heat by operating at low power levels and removing the sulfate utilizing r sactor water clean-up, c. Conclusion

..

When the chemistry Action Level 3 for sulfates was exceeded, plant management

temporarily changed GP-12 to avert the required Unit 1 shutdown. Although operation at j low power with the high sulfate levels did result in some degradation of vessel l

components, this was the industry accepted method for returning chemistry levels to

'

acceptable values. It appears that the procedure was changed in a timely manner; however, the process used was inappropriate since the change involved an intent change to the procedure. For this instance, safety consequences were low. Technical Specification 6.8.3 requires, in part, that temporary changes may be made to procedures of Specification 6.8.1 provided the intent of the original procedure is not altered. Technical

'

Specification 6.8.1 procedures include General Plant Operating Procedures for Power i Operation. These requirements assure'that appropriate safety reviews of intent changes j are conducted prior to implementation of the change. On September 23,1995,a

'

temporary change was made to a General Plant Operating Procedure, which altered the i

j

- - - - , .-. -- - .

.., .. . - _ - , , . .. --

..

'

I 6 intent of the original procedure. Specifically, GP-12 was changed to no longer require an immediate plant shutdown when Action Level 3 was exceeded, and a prior safety review of this change was not conducted. This is a violation. (VIO 50-352/97-03-03)

04 Operator Knowledge and Performance 04.1 Operator Restoration of the Reactor Water Clean-uo System - Unit 1 a. Inspection Scone (71707)

On May 1, the inspector observed the Unit 1 operators warm-up and place the reactor i water clean-up (RWCU) system in service following a two day outage to repair various l system leaks. The inspector reviewed procedures S44.1.G, Fill / Vent Unit 1 RWCU l Recirculation Pump, and S44.7.C, RWCU Delayed Hot Startup, and discussed the evolution I with the operators and the system manager.

b. Observations and Findinas The RO performing the evolution in the control room demonstrated an in-depth system knowledge and awareness of the impact the evolution had to system components and to reactor thermal power. The RO ensured that the procedures used were correct for the activity and reviewed them to understand completely what was being done. The evolution contained several infrequently performed steps, including agitated venting of the pump seal cooling water piping, system pressurization (which could cause an automatic high differential flow isolation), warming of the RWCU pump casing, and gravity blowdown to the main condenser. The RO responded knowledgeably to the inspector's questions concerning the objective of certain procedural steps and to the impact of the step on the plant.

The inspector noted that the RO's control was excellent during RWCU system pressurization and while establishing gravity blowdown. The RO established a slow pressurization rate by cracking open the upstream isolation valve, so that the high differential flow isolation was not actuated. Further, while establishing gravity blowdown, the RO continuously monitored the RWCU temperature effects on core thermal power because of the changing RWCU temperature inputs to the reactor heat balance. The RO made necessary corrections to recirculation flow to correct power.

The procedures were adequate for the evolution, although procedure S44.7.C, was temporarily changed (TC) to tailor it for the evolution. The system manager had to customize the system restoration plan and TC the procedures because a detailed procedure for returning the RWCU system to service while at power does not exist. The delayed startup procedure is written for reactor coolant temperatures greater than 200 degrees l

with the system out of service for periods greater than one hour. The inspector also noted several procedure steps that contained more than one action per step, which is considered

,

to be a poor human factor practice. The inspector, however, did not observe any difficulty

!

on the part of the RO in performing the procedure. The system manager recognized the weaknesses in the RWCU procedures after discussions with the inspector and initiated a revision to incorporate the lessons learned.

l

- - . - - -.- _--- - -

_

-- - _ _ _ _ - - . - - -

l l ..

i ..

! 7 c. Conclusions

!

l The operator performance in restoring the RWCU system to service following the outage l was excellent. Procedures, althmagh adequate, were not specific to the evolution, needed to be temporarily changed and were not well human factored in that they contained more l than one action per step, ll. Maintenance M1 Conduct of Maintenance M 1.1 General Comments a. inspection Scone (62707) '

The inspectors observed all or portions of the following work activities:

e D22 Emergency Diesel Generator (EDG) 18 Month Inspection e 1C Reactor Feed Pump (RFP) Discharge Check Valve Repair - Unit 1 e 1B Control Rod Drive Pump Repair - Unit 1 e Replacement of the Unit 1 Division C Safety-Related Battery e Alterex Bearing Replacement - Unit 2 e Alterex Service Water Leak and Neutral Bushing Repair - Unit 1 l

b. Observations and Findinas '

i On April 12, Nuclear Maintenance Division (NMD) technicians successfully repaired a j hydrogen leak at the C phase neutral bushing on the Unit 1 generator. The bushing was j leaking at a rate that required operators to add hydrogen up to twice per shift. The  ;

bushing will be permanently replaced during the upcoming refueling outage (1R07). l

!

Operations personnel identified a series of pin hole leaks on an 18 inch section of the I service water piping vent line to the Unit 1 Alterex cooling system. A temporary clamp 1 and catch containment were applied to reduce and collect the leakage until the piping is replaced, planned for the next extended outage.

During the Unit 1 load drop for the generator bushing repair, the 1C RFP discharge check valve experienced excessive chattering causing flow oscillations. NMD personnel and site engineering disassembled and inspected the valve internals to determine the cause of the valve chatter. The inspection determined that the linkage arm had rotated out of the correct orientation; the discrepancy was corrected. After inspecting the other RFP check valve linkages, NMD personnel found that the 1 A RFP discharge check valve linkage had also rotated. However, engineering decided that the present orientation had not affected the operation of the check valve and, thAefore, did not mandate an immediate repair. The Unit 2 RFP check valve linkages were operating satisfactorily and verified to be correct l during the recent refueling outage.

l

.-- __ _ _ _ _ _ _ - _ - - -

. _ . _ _ _ _ _ . . . _ . . _ . . _ . _ - . . . . . _ - . _ _ _ _ . . _ ~ . . _ . . _ . _ _ _ _ _ _ _ _

i .-

.

i

.i

.

..

i I

During the week of March 31, NMD technicians repaired the Unit 2 Alterex flexible ,

coupling. 'he coupling was identified to be leaking excessive amounts of grec=e which

'

was due to cutting a replacement gasket incorrectly during refueling outage 2R04. During

activities to replace the gasket, technicians identified several Alterex alignment pro)lems.

'

NMD performed extensive work to ensure the correct alignment of the Alterex and main ,

generator.

,

Due to recurring vibration problems at the Alterex bearing, an engineering task force

, determined that several factors were contributing to the vibration problems. The Alterex

! bearings replaced during 2RO4 were machined to be cylindrical instead of elliptical, as

! designed. Further, the shims used to align the Alterex were not providing adequate t

support. During the turbine outage that occurred the week of April 30, the Alterex was

completely disassembled and rebuilt using the appropriate bearings and enhanced shim

package. After the turbine was returned to service, the vibration data indicated a

i substantial decrease in vibration levels. -

!  !

1 i j c. Conclusions  ;

i

'

The two turbine outages to repair the main generator flexible coupling and Alterex bearing were well planned and implemented. The turbine vibration data following the outages indicated a substantial decrease in vibration levels.

M1.2 D22 EDG 18 Month Overhaul a. Insoection Scope (62707)

From April 7 through 14, the maintenance staff performed the 18 month inspection, 1 testing, and general maintenance on the D22 EDG. The scope of the outage included the replacement of the carbon steel exhaust stack with a stainless steel stack, and chemical flushing of the three skid-mounted heat exchangers. The inspectors reviewed the work permits, procedures, and the engineering documentation for the' stack replacement and chemical cleaning. The inspector also observed portions of the maintenance work and i post-maintenance testing.

' b. Observations and Findinos The maintenance technicians performed the outage well. The inspector observed that work activities were well controlled, with good housekeeping practices implemented. The inspector noted that maintenance technicians exercised good foreign material exclusion (FME) measures and that good FME practices were implemented and followed. The technicians were qualified and performed activities in accordance with approved procedures.

The maintenance team cleaned and inspected the tubes in the lube oil, jacket water, and '

combustion air heat exchangers during the outage. The engineering staff approved a chemical process for tube cleaning. The purpose of using a chemical process was to determine if the chemical process would be a better process than the hydrolazing process used until now. The results of the post chemical cleaning thermal performance test

.- ._. -. . - .- - -- - - - - . - . - . _ .

- - , -- --.-.-.- - - _ - _ _ _ - .- - ._ - _ .. - -. -.-

.-

,

l l

.

revealed that there was no discernible difference between the chemical and hydrolazing

processes.

! Preventing heat exchanger fouling continues to be a key focus for improving the EDG performance at Limerick. In March, during the D14 EDG inspection, an action request l addressed a potentially serious concern regarding 55 of 188 tubes in the tube oil heat i

exchanger being clogged with rust debris and mud. A special procedure and nonconformance report established ' methods to map and inspect the affected heat  !

exchanger areas and required documentation of "as-found" conditions. As a result of this 1 increased surveillance, the engineering staff identified that several reported clogs were l actually due to a sag in the heat exchanger tube. The tube remained full of murky water j and appeared to be blocked since a light placed at one end did not shine through to the other end. Pushing a piece of tygon tube through each heat exchanger tube found that the tubes contained murky ESW, but no debris or silt. The visual inspection showed that a sag in the tube created a visual illusion of blockage. These results were further confirmed -5 during the D22 inspection.

As a result of the D14 findings, all EDG heat exchangers are being cleaned prior to July to t assure maximum performance and efficiency. The engineering staff is benchmarking heat exchanger thermal performance data to determine the appropriate cleaning frequency of the EDG heat exchangers. Further, valuable experience and data are being gained in establishing proactive measures to prevent heat exchanger fouling.

During the outage, the first of eight type 316 stainless steel exhaust stacks was installed.

This modification was developed as a long term repair due to three previous and separate EDG crankcase overpressure events. The overpressure events had been caused by partial blockage of the exhaust pipe by laminar corrosion flaking off the piping wall and impeding the exhaust flow, creating a backpressure. The engineering change request (ECR) and

. safety evaluation were complete, addressing the corrosion issue and material change.

t The D22 EDG performed well during the post-maintenance surveillance test. Following the completion of all outage work, operators performed a modified D22 EDG Run-in test (RT-6-

092-312 2) which ran the engine at various output levels. Following this EDG run, the l D22 EDG 24-hour Endurance Test (ST-6-092-112-2) was performed with satisfactory results.

l The inspector noted that the general material condition of the EDGs is improving. Oil leaks are being identified and addressed. Programs to paint and improve habitability of the EDG I lcubicals are effective. Technicians are upgrading the lighting in the EDG area with higher lumen emitting bulbs and the addition of several new florescent fixtures.

! c. Conclusions l L l l' The inspectors concluded that the Limerick staff performed the 18 month inspection and L  : maintenance of the D22 EDG well.

!

i-l

'

,

l l

L _

_ _ _ _ ._ _ . . _ ._ . _ _ _ _ . _ _ _ _ . _ _ _ _ . _ . _ _ . . _ . _ _ _ _ _ _ _ . _ . - .

..

. . .

M1.3 Division 1C Safeauards Batterv Reolacement - Unit 1 a. insoection Scoce (62707)

Beginning on May 2, I&C technicians replaced all 60 cells of the Unit 1, Division 3, ,

Safeguards Battery. The work was performed on-line and was completed on May 3. The j

. inspectors reviewed the work authorization package and observed portions of the work i activities.

b. Observations and Findinas The battery replacement was a pre-planned activity and not the result of any battery cell failures, which have been experienced previously. The battery replacement was part of the Battery !mprovement Plan which is a proactive initiative to improve the overall performance of the Safeguards DC system. 1 The technicians replaced the battery cells within two days, over four shifts, in four groups l of fifteen cells. A partial surveillance test verifying the operability of the new cells was l

performed after replacing each group. This replacement methodology prevented the battery from being inoperable for extended periods of time.

The technicians demonstrated that they were knowledgeable and experienced in on-line battery replacement by using safe work practices. Maintenance supervision was present during the activity and the appropriate work orders and procedures were used.

The inspector noted that the portable battery charger did not have "Q" rated fuses to provide electrical separation between a "Q" rated and non "Q" rated system. The l&C supervisor was aware of the condition and stated that the battery.was declared inoperable whenever the portable battery charger was connected. The inspector verified that this action occurred.

Two cells from the old battery and one cell from the new battery are being analyzed at the Corporate Laboratory at Valley Forge. The results will be included in the history file being maintained by the Lab, with the intent of building a database of representative battery cell characteristics. An improved testing and preventive maintenance program will be developed based on this database, c. Conclusiong The inspector determined that the battery replacement work was performed safely and met the requirements of the Technical Specifications.

M1.4 General Comments on Surveillance Activities (61726)

The inspectors observed selected surveillance tests to determine whether approved procedures were used, details were adequate, test instrumentation was properly calibrated and used, technical specifications were satisfied, testing was performed by knowledgeable personnel, and test results satisfied acceptance criteria or were properly dispositioned.

-- - . - - - - - . . - - . . - . . - - - . - -. -

..

..

The inspectors observed portions of the following surveillance activities:

e Reactor Enclosure Secondary Containment Integrity Verification - Unit 1 e Standby Liquid Control Line-up Verification - Unit 2 e Reactor Enclosure Secondary Containment Integrity Verification - Unit 2 e Remote Shutdown System ESW and RHRSW Operability Test (partial)

e Division 3 Weekly Battery Test - Unit 1 e B RHR Pump, Valve and Flow - Unit 1 o D12, D22, D23, D24 Monthly Tests Observed surveillance tests were conducted well using approved procedures, and were completed with satisfactory results. Communications between the various work and support groups were good, and supervisor oversight was good.

M1.5 Overdue Surveillance Tests l a. Inspection Scone (37551. 61726)

The inspector reviewed a list of overdue surveillances to determine if required surveillance tests (STs) were being properly controlled and tracked. The inspector discussed three overdue STs with appropriate plant personnel to determine their safety significance, and to determine the present status of the testing.

,

b. Observations and Findings

!

! The list of. overdue STs contained approximately 40 surveillances which had exceeded the

! allowable time period, and could not be completed due to a variety of reasons; l- compensatory or other actions had been taken as appropriate to ensure continued l compliance. A large number of the tests were associated with the hydrogen water chemistry modification, which has not been fully implemented.

The day after discussing the list with outage management personnel, the inspector became - !

,

aware that one of the tests on the list was being performed late, in that it should have L been performed months earlier when the system was fully restored. For this instance, a

! service water valve had been returned to service, without testing circuit contacts associated with the remote shutdown panel. A note had been added to the clearance for the valve, in April,1996, directing performance of the surveillance test when the clearance i

'

was removed. The valve was declared operable on January 30,1997, and the missed surveillance was identified on April 29,1997; the surveillance test was successfully completed on April 29,1997. Through discussions with engineering and licensing personnel, and review of the appropriate circuit diagrams, the inspector determined that the valve was fully operable after restoration on January 30,1997. The circuit contacts which were not tested were not required for valve operability; their purpose was only to

, ~ provide a design enhancement so the system could be tested without impacting plant.

,

operation. Engineering personnel plan to revise the surveillance test procedure to indicate

! that testing of these contacts is not required for valve operability; other contacts will be l checked for the same corrective action. However, this example exhibited an apparent i

- . .,.- - -

_

,

i .

l l

!

12 l l

weakness in that the valve testing was missed in January 1997, when the valve was

! returned to service.

Since the above described surveillance test was missed and not identified for three months, even though it was listed on the overdue surveillance test list, the inspector reviewed the other surveillances on the list to ensure that no other problems existed. The inspector identified two other concerns associated with surveillances listed as overdue.

One test was associated with fire damper testing, and the other was associated with the D13 EDG gauges.

The overdue surveillance associated with the fire dampers was a test to inspect the integrity of the fusible links and other general inspections of the two standby gas treatment system HVAC dampers. Compensatory actions were implemented as required once the surveillance became overdue. The surveillance was not performed prior to the due date because operations personnel had questioned the method of performance; the test required a visual inspection of the damper area inside the duct work. This involved opening an access door associated with the standby gas treatment system, which could adversely affect its ability to perform its safety related function. This concern had been evaluated by engineering personnel in 1991, had been reviewed by PORC, and the surveillance test procedure was changed to incorporate administrative controls for the activity. A technical specification 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> action statement for secondary containment was entered each time an access door was opened.

The in:: octor reviewed the engineering evaluation for the dampers, the PORC minutes which . :/iewed the evaluation, the procedure for performing the test, and discussed the issues with engineering personnel. The procedure appeared to cover all of the administrative guidelines reviewed by engineering and PORC: operator log book entries documented that the appropriate action statement was entered each time the door was opened during a performance of the test in 1994, when it was last performed. Engineering personnel indicated that although the test could be performed now, and should have been performed earlier, they wanted to perform a more thorough review of the issue before the test's next performance. The inspector noted that this surveillance test had been rescheduled several times during the past year, and that ultimately plant management involvement was necessary to ensure that this test was promptly addressed.

The overdue uveillance associated with the emergency diesel generator concerned various instrumentation that needed repairs. The inspector observed that the instrumentation was repaired, but the paperwork was not yet closed out. However, the surveillance coordinator identified that the D13 fuel oil day tank levelindicator had been out of calibration for a number of months, and operators had been using it to meet the technical specification requirement for verifying minimum tank level, investigation revealed that the level was verified by alternate means; a level switch brings in a low level alarm l prior to reaching the technical specification limit, and no alarms were received during the i

time period in question. Additionally, the level indicator, which the operators were using, was out of calibration in the conservative direction such that the indication was lower than the actual level. Operators were not aware that the level indicator was out of calibration due to poor communications between the work groups involved, and the indicator was not marked with an equipment trouble tag.

-. .. . -- --.

_

l l

..

..

]

c. Conclusion l l

The inspector concluded that the late performance of the service water valve circuit !

contact test was weak, but there were no safety consequences associated with it, since the valve was always operable after restoration. The other two surveillances reviewed exhibited delays in resolution due to poor communications and tracking. Corrective actions are being evaluated. ,

l M3 Maintenance Procedures and Documentation  !

M 3.1 Post LOCA H2 Recombiner Flow Calibration incorrect at Unit 1 a. Inspection Scone (62707)

Two Fix-It-Now (FIN) team technicians identified that the primary containment hydrogen l recombiner flow loops, FT-DO-101 A & B, at Unit 1 were miscalibrated. Plant management informed the inspector that both hydrogen recombiners at both units could be potentially inoperable. The inspector reviewed the issue and the non-conformance report (NCR), and discussed the issue with the appropriate plant staff, b. Observations and Findinas l

During post maintenance testing for the flow recorders replaced on the primary l containment hydrogen recombiner, a FIN team l&C technician experienced a square root converter acting erroneously. He reviewed surveillance tests ST-2-057 406-1(2) and ST-2- j 057-407-1(2), Primary Containment Hydrogen Recombiner Channel A(B) Calibration, and i discovered that a difference in the calibration range for the recombiner flow loops existed I between the two units. The technician determined that the Unit 1 flow transmitters were j miscalibrated at 0-15 inches water column ("w.c.) rather than 0-12" w.c. as at Unit 2. i l

The system manager determined that the calibration mismatch resulted in a less than nominal error in flow. With the flow error, the recombiner would still be operated within the design limits and was operable. Operability was based on the ability to maintain gas temperature greater than 1150 degrees F to support the recombination process. A review of past surveillance tests indicated that the temperatures were satisfactory from the time that the instruments were originally miscalibrated. Also, the evaluation indicated that the greater flow, which resulted from the miscalibration, maintained oxygen levels less than five percent and was conservative.

The I&C technicians recalibrated the Unit 1 flow transmitters to 0-12" w.c. and revised the surveillance procedures for the correct limits as corrective actions. The cause for the calibration difference was due to installing a different flow transmitter during construction at Unit 2. Subsequently, the flow transmitter was replaced at Unit 1 with the similar type, i but the calibration parameters were not transposed to the Unit 1 procedure.

l

!

l l

.. .-. - - -.. - - _._ _ - - _- -. - . - . - - - - -._ -_ _- - _.

..

L ..

c. Conclusion U

The technicians demonstrated an excellent questioning attitude which led to identifying the discrepancy in the parameters between the two units.

M3.2 Reactor Enclosure Secondarv Containment Intearity Verification

.

a. Insoection Scope (61726)

!

,

The inspector observed the performance of surveillance test ST-6-076-360-1, Reactor

!

Enclosure Secondary Containment integrity Verification. During the performance of the verification, the inspector raised several concerns. These concerns were discussed with various Operations Department managers.

b. ' Observations and Findinas .

l The surveillance test satisfies TS surveillance requirement 4.6.5.1.1.b, which verifies that

'

penetrations and access paths into the reactor enclosure secondary containment are closed

'

and sealed. The operations department method for performing a portion of the ST was to refer to the Locked Valve Log and LCO log to identify possible secondary containment breaches.

!- Several floor drains located in the refuel area and air supply fan room were verified to be plugged and locked. Many of these drains are located in radiological areas and, due to ALARA concerns, the operator performing the ST may verify the condition of the drains using the Locked Valve Log. The inspector raised a potential concern, however, in that

.

reviewing the Locked Valve Log may not sufficiently verify the condition of these drains.

Administrative Procedure A-C 8, Control of Locked Valves and Devices, allows for two conditions when a locked valve does not have to be entered into the log: when locked valves and devices are manipulated by 1) approved tests (e.g., Surveillance Test, Special Procedures) that include documentation for restoration and Independent Verification; and 2)

clearances that have documentation for restoration and independent Verification.

Operations management initiated a PEP review of the inspector's concern. The evaluation focused on detern.ining if the review of the locked valve log in the performance of the ST was adequate to meet the requirements of the TS or if a non-compliance situation existed.

Subsequent to the PEP, management changed the surveillance procedure to include review of the Barrier Breach Log (A-C-134), active clearances, and the LCO and Potential LCO Logs to supplement the locked valve log review. These changes were made to enhance the surveillance test. The PEP concluded that the previous method of verification

- questioned by the inspector was adequate and that the method of performing the ST was in compliance with TS.

-The inspector observed the performance of the ST in May and verified that the procedure changes were in place and performed adequately. A joint meeting is planned with Peach Bottom for June 1997, to discuss the use of the Locked Valve Log to determine a common approach to maintaining the TS criteria in a common method.

_ -- . - _ _ - - _

.__ __ ._. ___ _ __ _ _ _ . _ _ _ _ _ _ . - _ . _

. ._

__ .

..

.

c. Conclusions Operations management enhanced the method of performing the Reactor Enclosure Secondary Containment Integrity Verification, after evaluating several concerns raised by the inspector. The issue will be discussed with personnel from the Peach Bottom Station, and will remain unresolved as the ST's conformance with TSs until after the discussion.

(URI 50-352, 353/97-03-01)

111. Enalneerina l E2 Engineering Support of Facilities and Equipment E2.1 Inocerable Unit 1 Steam Floodina Damcers a. Insoection Scope (37551)

The inspector reviewed the results of the testing performed on six Unit 1 steam flooding dampers, which were not previously tested along with all other dampers in February 1997, due to the dampers being inaccessible at the time. The root causes of failures and corrective actions were discussed with engineering personnel to understand why these l dampers were evaluated and considered operable in February 1S97, since four of the six

! failed in April 1997.

b. Observatiens and Findinas in February 1997, plant personnel identified a number of steam flooding dampers which failed to operate properly due to binding (see NRC Integrated Inspection Report 50-352/97-01, 50-353/97-01). Six Unit 1 dampers were not tested during the February time frame due to the inaccessible location of the dampers; they are located in high radiological dose l

areas. Plant engineering personnel evaluated the six dampers and concluded that they l were operable, based on: all six dampers were last inspected and satisfactorily tested in l

February 1996; records indicated a history of successful testing for all 6 dampers; the overall damper failure rate identified in February 1997, was low; and only one darnper, which had previously been satisfactorily tested in the prior 3 years, failed. Management committed to testing the six untested dampers during the next opportunity, when plant power was reduced so that the areas were no longer inaccessible due to high radiological conditions.

On April 12, Unit 1 power was reduced sufficiently to allow testing of the six dampers.

Four of the six dampers failed to actuate, due to excessive friction in the bushing located between the solenoid and the damper linkage. This failure cause is the same as that

'

identified for the dampers in February 1997; corrective actions were also the same as those in February, in that all six were cleaned and lubricated, and the bushings were replaced. However, the inspector was concerned that four of the dampers failed, when

'

they were evaluated as operable in February 1997. Through discussions with engineering supervision, the inspector determined that there were differences between the four dampers which failed in April and the dampers which failed in February. Some dried l

. - - - . - - - - - - . . - - - - -

. - .-.- - - -.-- - .--_. .-.-

.

l l .,

l 16 grease was found on the linkages of the six dampers in April, but no grease at all on the dampers in February; apparently the environment where the dampers are located is more harsh than other areas. Additionally, the dampers were older than most of the other dampers. Grease is not supposed to be on the linkage, lubrication is with a light lubricant.

c. Conclusion -

,

The inspector concluded that the original engineering analysis for the six untested dampers did not address the potential for wrong lubricant as a failure mechanism, since no dried

'

grease was identified on any of the damper linkages in February 1997. Therefore, that cause would not have been reasonably expected for the four dampers in April. The failure

! of the four dampers is another example of the non-cited violation described in NRC Integrated Inspection Report 50-352/97-01, 50-353/97-01.

l E8.1 '(Closed) LER 1-97-003, Revision 1, Dearaded Back Pressure Dampers Needed for j Pioe Ruoture Mitiaation Result in Ooeration Outside Desian Basis

'

l This LER was originally reviewed in NRC Integrated Inspection Report 50-352/97-01, 50-

! 353/97-01. This revision provided additional information concerning the testing of Unit 1

dampsrs, which were inaccessible and therefore not tested in February. Review of these l dampers' tests is in section E2.1 of this inspection report.

E8.2 (Closed) LER 1-96-021, Revision 1. Failure to Provide Sufficient Repair Actions Needed to Achieve Cold Shutdown for Fire Safe Shutdown Capability (90712)

The original revision for this LER was reviewed in NRC Integrated Inspectiun Report 50-352/96-10, 50-353/96-10. This revision provided additional background information and corrective actions.

IV. Plant Support

.

R1 Radiological Protection and Chemistry (RP&C) Controls

R1.1 Chemistry Controls a. Insoection Scoce (71750)

The inspector reviewed circumstances surrounding two previous chemistry-related activities: 1) in September 1995, after the Unit 1 safety relief valve (SRV) stuck open,

" dirty" suppression pool water was directed into the reactor vessel; and 2) the diversion of unfiltered reactor water to the condensate storage tank (CST),in March 1993.

I b. ' Observations and Findinas l

! 1) Did " dirty" suppression pool water get injected into the Unit 1 reactor in September 1995? Was that appropriate at the time?

l

_ _ __

__ __ . .- - - --- - .. . .. - _ - - - -.-

..

i

.,

The time period in question for this chemistry-related activity was after the Unit 1 SRV stuck open on September 11,1995. The two NRC inspectors in the control room shortly after the SRV lift, who were there for several hours, were questioned and stated that suppression pool water was not directly injected into the reactor.

The shift manager, who was on duty during and after the event, and operations and chemistry management, indicated that suppression pool water was not directly injected into the reactor. Review of chemistry and operation logs did not identify any instances where suppression pool water was directly injected into the reactor.

However, suppression pool water did make its way to the reactor. After the SRV lifted, when the suppression pool level increased, operations personnel pumped suppression pool water to the hotwell to lower the suppression pool level. This is a normal and acceptable operation, and the water is normally sent through condensate filter /demineralizers prior to being pumped to the reactor. The inspector did not find any log entries indicating that the condensate filter /demineralizers were bypassed during the time period in question (from the day before the event through the day after the event), but even if the filter /demineralizers were bypassed, putting this water into the vessel is not a problem unless it adversely affects primary chemistry. The inspector found no log entries or other indications that primary chemistry was adversely affected.

l l 2) Was reactor water let down, unfiltered, to the CST in early 1993? If so, were ,

! adequate precautions taken before, during, and after the discharge? What is the l I appropriateness and compliance of this activity? What was the impact on the control rod drive (CRD) filters later?

i Chemistry and control room operator logs indicate that near the end of the Unit 2 l refueling outage, on March 4,1993, RWCU was lined up to the CST without the l filter /demineralizers in service. No records were found indicating that this happened

! in January or February 1993. A chemistry log entry, for March 4,1993, indicated l that the chemistry data activity of 3.22 N4 meets CST activity limits (N4 means ten

'

to the minus fourth power). Chemistry indicated in the logs that they expressed concern to the shift manager about the possibility of hot particles going to the CST through an unfiltered path. Chemistry said that the shift manager made the decision l to use the CST until radwaste could take the water, at which time, the lineup would be made to radwaste.

The operations logs indicate that the RWCU to the CST buy-in was via chemistry, and HPs had posted the areas in the reactor building (elevations 283 & 201). The shift manager's log specifically states that HP and chemistry approved dumping RWCU to the CST without filter /demineralizers, that activity was low enough for this to be acceptable, and that HP posted and surveyed appropriate areas.

The current Chemistry Manager said that this practice was acceptable; HP

supervision said this was acceptable and has been done a number of times, though he indicated that it is important for HP involvement especially since the Unit 2 CST i is fairly near the protected area (PA) boundary fence; the shift manager didn't

.

remember this specific instance, but said this has been done a number of times and

1 ..

..

he believed that it was acceptable at that time also, based on review of the log entries.

All plant personnel interviewed indicated that the process is acceptable and is a radiological concern only. It appears that proper radiological precautions were taken for this instance. The concern with the CRD filters is a radiological one also, since radiologically contaminated water (CRD pumps take suction from the CST) would make the CRD filters radiologically hot, which would affect the dose received during replacement of the filters.

Limerick's administrative limit for activity in the CST is (and has been for a long time) 1 N3, per chemistry procedure CH-1010, Appendix A. This limit was not exceeded on March 4,1993. In general, since this is an administrative limit, it may be acceptable to exceed this, but this would have to be reviewed on a case-by-case basis, c. Conclusion Suppression pool water was not directly injected into the reactor in September 1995.

Suppression pool water was pumped to the hotwell, and then to the recctor, but this was acceptable and the water was probably cleaned first. No noncompliance was identified.

Although reactor water was not found to be let down to the CST in January 1993, it was done in early March 1993. However, log entries indicate that appropriate and adequate precautions were taken, and the appropriate personnel were involved. Also, this activity has been properly performed a number of times, without problems. Activity levels were below the administrative limits for the CST, and the CRD filters were affected by the water, but this is a radiological concern, and appropriate precautions are taken during replacement activities. Chemistry personnel searched to see if the administrative limit for the CST was exceeded at another time, but did not identify any instances. No noncompliance was identified.

R5 Staff Training and Qualification in RP&C R5.1 Chemistry Manaaer Qualifications a. Insoection Scooe (71750)

The inspector reviewed past chemistry managers' qualifications to determine if the chemistry managers fully met qualification requirements throughout the past five years, b. Observations and Findinas

.

Limerick Technical Specification 6.3.1 requires that each member of the unit staff shall l meet or exceed the miNmum qualifications of ANSI /ANS 3.1-1978. The Limerick UFSAR l presently requires, and required in 1992, that the Manager - Chemistry (formerly Senior Chemist) shall meet or exceed ANSI /AN3 3.1-1978,4.4.3. The Limerick UFSAR required, from November 1993 through April 1996, that the Manager - Chemistry and the two

~

..

..

Managers - Plant Chemistry shall meet or exceed ANSI /ANS 3.1-1978, 4.4.3. ANSI /ANS 3.1-1978, 4.4.3 requires that the responsible person for chemistry and radiochemistry shall have five years experience in chemistry of which one year shall be in radiochemistry at an operating nuclear power plant.

From May 1993 until May 1994, PECO management indicated that the Chemistry Manager did not fully meet the ANSI requirements for the position. The Manager of Chemistry and the Plant Manager indicated that they informed the Primary Chemistry Manager that until the Manger of Chemistry met the qualifications (approximately one year), the Primary Chemistry Manager was the person responsible for meeting the requirements. From November 1993, until April 1994, PECO management indicated that they did not have two Plant Chemistry Managers who were fully ANSI qualified; they did have one fully ANSI qualified Plant Chemistry Manager. PECO management indicated that prior to early in 1995, they did not formally control and document how activities were conducted to ensure that a fully ANSI qualified individual was always appropriately involved. However, they indicated that they had at least three individuals in the chemistry department (one manager, and two supervisors) who were fully ANSI qualified, and were appropriately involved in activities requiring ANSI qualifications. PECO management indicated they believed the Chemistry Manager was ANSI qualified in May 1994; throughout the time period in question, one of the Plant Chemistry Managers was always fully ANSI qualified, and the third individual was fully ANSI qualified in April 1994.

c. Conclusion The Chemistry Manager and the BOP Chemistry Manager did not meet the minimum requirements for experience for some period of time after appointment to the positions; however, other chemistry personnel met the requirements and currently all chemistry management meet the requirements. In July 1995, a procedure was implemented (A-C-031, Nuclear Staff Qualification Requirements and Organizational Structure Control), which provides administrative controls for controlling and maintaining the qualification level for the Nuclear Generation Group. The failure to meet the minimum experience requirements for two chemistry management personnel is a violation of Technical Specification 6.3.1.

This failure is being treated as a non-cited violation, consistent with Section IV of the enforcement policy.

R8 Miscellaneous issues R8,1 (Closed) Unresolved item 50-352,353/96-10-04. Loss of Control of Master Kevs This unresolved item concerned an instance where PECO Energy personnel identified the failure to properly control several master keys for several years; the keys would allow access to numerous plant areas, including vital areas and locked high radiation areas. The security aspects of this issue are covered in Section S.8 of this inspection report. The j -radiological aspects of the issue were reviewed in NRC Integrated Inspection Report 50-l 352/97-01, 50-353/97-01. Technical Specification 6.12.2 requires, in part, that areas j accessible to personnel with radiation levels such that a major portion of the body could receive in 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> a dose greater than 1000 mrems shall be provided with locked doors to

! prevent unauthorized entry, and the keys shall be maintained under the administrative

,

..

..

control of Shift Supervision on duty and/or the health physics supervision. Contrary to this, unauthorized keys to locked high radiation areas, not under the administrative control of Shift Supervision on duty and/or health physics supervision, were available to personnel (fire protection personnel) since about 1993 through about November 1996. This is a violation of Technical Specification 6.12.2. (VIO 50-352,353/97-03-02)

S8 Miscellaneous Security and Safeguards issues a. Inspection Scone (81700)

On April 22,1997, a regional security specialist was onsite to review the licensee's investigation into the improper control of the vital and locked high radiation area keys.

b. Observations and Findinas The inspector's review of the licensee's investigation disclosed the following:

• In September 1996, the licensee identified that the corporate locksmith had made unauthorized copies of a grand master (GM) key from 1993-1996 t.ad provided those keys to several PECO employees to facilitate them in carrying out their duties at non-nuclear facilities. However, the keys also provided access to certain locked doors at the nuclear facilities. This was not realized until sometime later. An investigation was initiated by corporate security to review the scope of the problem.

• In January 1997, the corporate investigators identified that the unauthorized GM keys had the potential to provide access to the licensee's nuclear facilities [ Peach Bottom (PB) and Limerick Generating Station (LGS)].

e On January 31,1997, the investigators confirmed that the unauthorized GM keys would provide access to locked high radiation areas at PB and LGS and to vital areas at PB. Access to vital areas (VAs) at LGS was not possible because the key cores on the doors to VAs that the unauthorized GM keys would have opened had been changed in July 1996.

• When the licensee confirmed that the unauthorized GM keys would unlock VA and locked high radiation doors, compensatory measures were implemented. The compensatory measures for the high radiation area door locks at LGS doors were suspended when it was thought that all unauthorized GM keys had been recovered.

However, the compensatory measures were reinstated after the discovery that there were additional GM keys in existence. Compensatory rneasures remained in effect until the lock cores were changed. Compensatory measures for the PB VAs and high radiation areas were not suspended and remained in effect until the VA key cores were changed to a core series that the unauthorized GM keys would not open.

• In addition, to implementing compensatory measures for high radiation and VA doors that the unauthorized GM keys could open, the licensee made a 1-hour

_

..

..

notification to the NRC on January 31,1997, in accordance with the provisions of 10 CFR 73.71, identifying the problem.

• The licensee performed a review of security and high radiation area control records and concluded that there were no unexplained VA door alarms during the periods that personnel who possessed the unauthorized GM keys were onsite, and that no high radiation exposures, which could have resulted from unauthorized entry in high radiation areas, were experienced by those individuals, e Key core change out of locked high radiation area doors was completed at LGS on February 2,1997, and at PB on February 6,1997. The key core change out of VA at PB was completed on February 3,1997.

• On February 28,1997, the licensee concluded that the 1-hour report made in accordance with 10 CFR 73.71 was not necessary and rescinded the report, c. Conclusions The inspector concluded the following regarding the security aspects of this matter.

e The licensee aggressively pursued the issue when corporate security identified that the unauthorized GM keys had the potential to provide access to VA and high radiation areas at the nuclear facilities and implemented timely and appropriate compensatory measures as required by the NRC-approved security plans.

e The inspector concurred with the licensee's determination that the 1-hour report made in accordance with 10 CFR 73.71 was not necessary and the decision to rescind the report was appropriate.

• Based on the licensee's investigation determination that the individuals who possessed GM keys did not know that the keys would open doors at the nuclear facilities and because the VA doors would have alarmed and would have been responded to by a security officer if an unauthorized key was used to access a VA, the unauthorized keys did not represent an unmonitored security vulnerability.

e The locksmith violated licensee procedures by making the unauthorized GM keys and distributing the keys to individuals to facilitate their access to licensee non-nuclear f acilities. However, since the individuals who received the keys did not know the keys would open doors at the nuclear facilities, and because the locksmith lacked an understanding of the regulatory requirements and impact on nuclear security controls, it would appear that he did not intentionally compromise regulatory requirements regarding control of keys for the nuclear facilities.

I

i

..

.

. ,

22 i l V. Maneaement Meetinas l

l X1 Exit Meeting Summary The inspector presented the inspection results to members of plant management at the conclusion of the inspection on May 30,1997. The plant manager acknowledged the inspectors' findings. The inspectors asked whether any materials examined during the inspection should be considered proprietary. No proprietary information was identified.

X2 Review of UFSAR Commitments  ;

l A recent discovery of a licensee operating their facility in a manner contrary to the UFSAR l description highlighted the need for a special focused review that compares plant practices, j

!

procedures and/or parameters to the UFSAR description. While performing the inspections I discussed in this report, the inspector reviewed the applicable portions of the UFSAR that l related to the areas inspected. The inspector verified that the UFSAR wording was i

consistent with the observed plant practices, procedure and/or parameters. l l l

I

l l

l l

!

!

l

4gr . . - - - + - * w

. . . - - . _ . . _ _ _ _ . . _ . _ . . _ . _ . _ _ _ . . _ _ . _ _ . _ _ . _ _ _ . . _ _ . _ . . _ _ _ . . _ . _ _ _

r

,.

j

,

..

INSPECTION PROCEDURES USED

IP 37551: 'Onsite Engineering IP 61726: Surveillance Observation IP 62707: Maintenance Observation IP 71707: Plant Operations IP 71750: Plant Support Activities IP 90712: In-office Review of Written Reports IP 90713: Review of Periodic and Special Reports .;

IP 93702: Prompt Onsite Response to Events at Operating Power-Reactors ITEMS OPENED, CLOSED, AND DISCUSSED

.

Opened 1 352,353/97-03-01 URI Performance of Reactor Enclosure Seconc , Containment Integrity Verification (M3.2)

Closed 352,353/97-03-02 VIO Loss of Control of Master Keys (R8.1)

50 352/97-03-03 VIO Inappropriate use of temporary procedure change (03.1)

352,353/96-10-04 URI Loss of Control of Master Keys (R8.1)

352,353/1-97-003 LER Degraded Back Pressure Dampers Needed for Pipe Rupture

!

Mitigation Result in Operation Outside Design Basis (E8.1)

'

352,353/1-96-021 LER Failure to Provide Sufficient Repair Actions Needed to Achieve Cold Shutdown for Fire Safe Shutdown Capability (E8.2)

50-352/97-03-03 VIO Inappropriate use of temporary procedure change (O3.1)

Discussed-None i

i

5 i

l l

!

I

. - - . . - , . _ _ . - . _ . . _ . _ , _ . , .. - - . . - -

i

..

l .

LIST OF ACRONYMS USED

.

! .

ALARA As Low As Reasunably Achievable CFR Code of Federal Regulations l

CRD Control Rod Drive j CRS Control Room Supervisor l CST Condensate Storage Tank-ECR Engineering Change Request EDG Emergency Diesel Generator  ;

EHC Electro-hydraulic Control i EO Equipment Operator

!

EPRI Electric Power Research Institute i ESF Engineered Safety Feature ESW Emergency Service Water ETT Equipment Trouble Tag FIN Fix It Now FME Foreign Material Exclusion GM Grand Master HP Health Physics HPCI High Pressure Coolant injection l HVAC Heating, Ventilation and Air Conditioning l IFl Inspection Follow-up Item ISEG Independent Safety Engineering Group I&C Instrumentation and Control LCO Limiting Condition For Operation LER- Licensee Event Report FIN Fix-It-Now

, LOCA Loss of Coolant Accident i LOOP Loss of Offsite Power

! NCV Non-Cited Violation NMD ' Nuclear Maintenance Division j i NQA Nuclear Quality Assurance l NRC Nuclear Regulatory Commission l- PEP Performance Enhancement Process l PORC Plant Operations Review Committee i QA Quality Assurance RFP Reactor Feed Pump RHR Residual Heat Removal RHRSW Residual Heat Removal Service Water

[ RO Reactor Operator l RP&C- Radiological Protection and Chemistry RPS Reactor Protection System RT Run-in Test RWCU Reactor Water Clean-up RWP Radiation Work Permit SRV Safety Relief Valve ST Surveillance Test TC Temporarily Changed i TCV Turbine Control Valve

TS Technical Specification j UFSAR Updated Final Safety Analysis Report

URI Unresolved Item ( VA Vital Areas VIO Violation a -- - ---

. _