IR 05000293/1992023

Safety Insp Rept 50-293/92-23 on 921020-1123.No Violations Noted.Major Areas Inspected:Plant Operations,Radiological Controls,Maint & Surveillance,Emergency Preparedness, Security & Safety Assessment & Quality Verification

ML20126L519
Person / Time
Site:	Pilgrim
Issue date:	12/29/1992
From:	Eugene Kelly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20126L506	List: IR 05000293/1992023 ML20126L503
References
50-293-92-23, NUDOCS 9301080057
Download: ML20126L519 (29)

Text

. _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _

.:

,

U. S. NUCLEAll itEGULATOltY COMMISSION ItEGION 1 Docket No.: 50-293 Iteport No.: 92-23 1.icensee: 130ston DJison Company 800 Ik)ylston Street ik)ston, Massachusetts 02199 Facility: Pilgrim Nuclear Power Station lecation: Plymouth, Massachusetts-Dates: October 20 - November 23,1992

'

Inspectors: J. Macdonald, Senior itesident inspector A. Cerne, llesident inspector D. Kern, llesident inspector E. King, Physical Securit > Ins ' r Approved by: M * -

E. KelIy, C f, itcactor Projec Section 3A Date Scope: llesident safety inspections during the midcycle outage in the areas of plant -

operations, radiological controls, maintenance and surveillance, emergency preparedness, security, safety assessment and quality verification, and engineering and technical support. Initiatives selected for inspection included a walkdown and functional evaluation of the salt service water (SSW) system, motor-operated valve maintenance, reactor vessel level instrumentation performance, and the continued -

implementation of the SSW system piping replacemen .

Inspections were performed on backshifts during October 22, 27 29 and November 2, 4, 5,12, 16-19, and 23. " Deep" backshift inspections were -

performed on October 24 from 7:00 am - 8:00 pm and on November 5 6 from 11:00 pm to 2:30 a Findings: Inspection results are summarized in the Executive Summary.

l 9301000007 921231 PDR ADDCK 05000293 g PDR <

p

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

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

r p

.

EXECUTIVE SUMMARY Pilgrim Inspectiun Report 50 293/92 23 Plant Operations The overall control of plant operations was good, and operators were cogninmt of maintenance and modifications in progress during the midcycle outage. Excellent lineup and tagging activities for salt service water system train separation maintained the plant in a safe configuratio ,

Radiological Controls

Radiological protection support for outage maintenance, particularly the use of the optical disk surrogate plant tour, was outstanding. Reviews of outage maintenance work packages were unsuccessful in identifying the high radiation levels in the radwaste demineralizer room, but detailed corrective actions were promptly implemente Maintenance and Surveillance A heat exchanger repair to stop a small coolant leak in the reactor water cleanup system was well planned and executed in accordance with ASME Code requirements. Other field activities '

were observed to be properly controlled and implemented by knowledgeable personnel. An aggressive motor-operated valve (MOV) maintenance schedule was accomplished during the midcycle maintenance outage. Several generic issues were appropriately addressed, which further improved valve reliability. An overall implementation schedule for MOV preventive maintenance remained under development due to pending revisions to the program's scop Emergency Preparedness Annual testing of the Prompt Alert and Notification System on November 19 successfully ,

demonstrated the licensee's capability to notify occupants of the ten mile emergency planning zone in the event of an emergenc Security Effective security measures were taken to compensate for any degraded conditions resulting from the midcycle outage work. iWCo promptly responded to potential problems identified by the NRC with access controls for equipment undergoing nullfication. The Fitness-for-Duty Program was found to be effectively implemente i

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

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

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

.

.,

EXECUTIVE SUMM ARY (CONTINUED) ,

Safety Assessment und Quality Verification Sound safety perspectives were demonstrated in the development of the midcycle outage schedule. The " system window" approach was effective in maximizing decay heat removal capability and electrical distribution system flexibility. Quality Assurance Department verification of operator tours has been effective, utilizing objective criteria to independently assess Operation Section self monitorin Engineering and Technical Support ilECo's issue Team continues to evaluate the reactor vessel level instrumentation performance i

in a ddiberate and technical fashion. External consultant expertise has been effectively utilize The presence of noncondensible gases in the reference legs was confirmed. The issue Team appropriately factored in Pilgrim specific experience with respect to llECo's response to Generic Letter 92 04, as well as to the generic considerations of the lloiling Water Reactor Owner's Group. Salt service water piping replacement activities continued, and were found to be well

'

planned and conservatively execu:c .

.

l l

,

s -e p -o--p- py- p-,--wwwy,-r--- - >,---t - - - - ++- 4- -

r we-- w-c-s-w- saww v- -

m -e- , - . , em- m -

t--,m -

ggrr--,- m -, r MM.-r >re--

___ _ _ _ _ _ - -__ .

.

.

TAllLE OF SUMMARY OF FACillTY ACTIVITils  !

)

At the start of the report period Pilgrim Nuclear Power Station was operating at approximately i 100% of rated powe l On October 23, 1992, operations personnel began the pro; css of deinerting the drywell in l preparation for the start of the scheduled 30 day midcycle surveillance testing and maintenance outag At 8:00 pm on October 23, control rann operators initiated the reactor power reduction. A thermal backwash of the main condenser was conducted at approximately 50% of rated power. On October 24,1992 at 5:40 am, the turbine generator was disconnected from the offsite distribution system. All control rmls were fully inserted at 10:06 am on October 24 and shortly after at 10:45 am the reactor maic scleet switch was placed in the SilUTDOWN l positio '

At 11:10 am on October 24, a spurious spike of two intermediate range monitors caused an automatic reactor protection system actuatlon. No control rod motion tesulted, as all rods had previously been inserted. Reactor vessel water level instrumentation " spiking" was experienced during reactor depressurization. The spiking began at approximately 350 psig and was similar .

to that experienced during previous reactor shutdown and depressuriration evolutions, i At 10:28 pm on October 24, an unplanned automatic primary containment isolation system Group I actuation occurred as the result of a spike in the reactor vessel water level

'

instrumentation associated with the "11" reference leg. At the time of the event, the reactor was -

essentially depressurized at about 2 psig and the reactor coolant system temperature was about [

165 depees fahrenheit. The main steam isolation valves had previously been closed during the normal course of reactor depressurization and cooldown, and were unaffected by the actuatio The appropriate main steam drain line and reactor sample line isolation valves repositioned to the closed position as designe Limiting outage testing and maintenance was completed on " A" train systems on October 30, 1992. Restoration of "A" train systems was completed and "II" train system outage testing and maintenance was initiated on November 1,1992. Limiting "ll" train activities were completed and the systems were restored to service on November 20,1992, Additional " A" train activities -

were conducted on November 20-22, 1992. Non-train limiting activities were completed on l November 23, 1992.

l-The reactor mode select switch was placed in the STARTUP position at 4:00 pm on November 23, 1992, following completion of final valve lineups and startup prerequisite At the

,

conclusion of the report period, the reactor was in the initial phase of power ascension.

l.

l l

l i

, . ,. .

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

i

-

)

!

.

2 1 FI, ANT OPER ATIONS (71707, 40500, 90712) Plant Operntions Review  :

.

The inspector observed plant operations during regular and backshift hours of the following l areas:

,

Control Room Fence Linc -

Reactor liullding (Protected Area) ,

Diesel Generator llullding Turbine lluilding >

Switchgear Rooms Screen llouse -

Security Facilities

,

Control room instruments were independently observed by NRC inspectors and found to be in '

correlation amongst channels, properly functioning and in conformance with Technical Specifications. Alarms received in the control room were reviewed and discussed with the operators. Operators were found cognirant of control board and plant conditions. Control room and shift manning were in accordance with Technical Specl0 cation requirements. Posting and i control of radiation, contamination, and high radiation areas were appropriate. Use of and - ,

compliance with radiation work permits and use of required personnel monitoring devices were con 0rme Plant housekeeping controls, including control of flammable and other hazardous materials, were observed to be appropriate. During plant tours, logs and records were reviewed to ensure compliance with station procedures, to determine if entries were correctly made, and to verify '

correct communication of equipment status. These records included various operating legs, turnover sheets, tagout, lifted lead and jumper logs, and were found to be satisfactorily maintaine .2 Drywell inspectinn

,

On October 24,1992, the inspector accompanied the operations section manager and members of the maintenance and technical staff on an initial drywell inspection with the reactor at rated pressure and prior to placement of the reactor mode select switch to shutdown.' Radiological-section personnel properly established the drywell control point. The radiation' work permit _ ,

briefing was thorough and accurately reDected survey data and permit precautions and^

requirements. Station security established the drywell access control point and maintained i positive control of personnel entering the drywell. The drywell inspection was comprehensiv Each drywell level was observed for system leakage; recirculation pumps were inspected for seal .

package leahage, oil levels, and general condition; and the undervessel area was inspected for ,

potential control _ rod drive leakage. All areas were observed to be in good material condition with no abnormal leakage noted. However, minor leakage from the "B" recirculation loop d_ischarge isolation valve, MO 202-511, valve body drain line was identined. The drain line is-0.75 inches in diameter and drains to the clean radwaste system via two 0.75 inch hand operated '

P

. - _ _ _ _ _ - - - - - - - --. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . ---_----_-_-___ ____ _ _ _ ______

,

l

!

l'

.

series globe valves (110-202-11311 and 110 20211411). During the outage, the hand operated  !

valves were replaced. The inspector reviewed the work package and discussed the valve replacement with involved maintenance personnel and identified no concern . Snit Service Water (SSW) System Walkdown (71710)

i in conjunction with the inspection of SSW piping replacement and system modification activities (see section 8.2), the inspector conducted a walkdown of the SSW system to include a review of component status, an examination of equipment condition and configuration, and an evaluation  ;

'

of the overall system lineup for the current PNPS operating mode. This inspection was '

conducted with the plant shutdown, the "11" SSW train of equipment undergoing modification, and the "A" SSW train operating in accordance with the Technical Specifications (TS). The _

inspector verined train isolation and proper valve lineup and control switch positioning (e.g.,

swing pump P-208C aligned for loss of power contigency) in the control room and spot checked the associated component tagging in the fiel During the examination of SSW equipment in the intake structure, plant auxiliary bays, newly constructed piping vault and the excavated SSW piping areas, the inspector checked material conditions and the impact of scaffolding and other outage work control items on the operation of the 'A" SSW train. The inspector also reviewed the accuracy of the current revision (E44)

of the SSW piping and instrumentation drawing (P&lD) no. M212, noting that the relevant plant i design change (PDC) information that was outstanding to this P&lD had been posted by document control personnel, included among the outstanding PDC's were the _ ongoing -

modi 0 cations to the hypo-chlorination system and SSW piping replacement activities, which the inspector confirmed to be in progress during the walkdown inspectio The inspector identified a minor P&lD discrepancy relative to the in line positioning of two thermowell, dual temperature elements (TE-6240/TE-3890 and TE-6241/TE-3891), in each case, the TE's were incorrectly represented to be upstream of the backwash piping connection to the reactor building closed cooling water (R11CCW) heat exchanger. However, the inspector determined that the "A" and "II" train piping isometric drawings accurately reflected the TE kications and that the P&lD crror was insignificant from an operational standpoint. . The licensee initiated a request to correct the minor discrepancies when P&lD M212 is next revised. The inspector also identined a missing condulet cover for the wiring of one TE. The licensee issued ,

a work request, reference tag no. 027613, to correct this dencienc I Additionally, the inspector examined the design and as-built configuration of certain SSW system -

pipe hangers in the intake structure. The inspector reviewed specific hanger drawing details, as well as isometric drawing no. JP-2915, and noted that certain discrepancies between the isometric drawing and the neid location of a few hangers, had already been identiSed by the licensee and were documented in problem report 92.0541. With respect to hanger 1129 1-23, the inspector questioned the orientation of a load carrying plate and its attachment welds and: ,

requested the design calculation set for this hanger. The inspector reviewed calculation no. SK-C-924(Q), as augmented by a licensee weld check verincation to confirm the structural adequacy

,

.

.

of the specific welds questioned by the inspector. These licensee calculations verified the load carrying capacity of the support welds with a substantial safety margin. The inspector also examined the sampled SSW piping supports for _ workmanship, material, member sizes, dimensions, tolerances and bolting practices. No nonconforming conditions were identined and the inspector had no further questions regarding hanger design or constructio OveraD, the inspector determined that the SSW system was built and was being maintained substantially in accordance with design and quality requirements. System operation during the current midcycle outage, was found to be in compliance with the Technical Specincation Acceptable work controls (e.g., safety and work request tagging, scaffolding use, maintenance work packages) in accord with procedural requirements, were noted. The inspector observed goal knowledge of system status and component availability and good interdepartmental communications in practice by the plant operators during inspection visits to the control room on several shifts. lixcept as noted above as minor discrepancies which are being corrected by the licensee, no additional concerns were identined d. iring this inspection walkdown of the SSW syste .0 R ADIGI,0GICAl, CONTitol S (71707)

The inspector reviewed radiological controls in place as well n the radiological conditions of selected areas of the plant, hianagement tours of the radiological control area (RCA) continued to be thorough and directed toward minimizing personnel exposure. Survey postings, radiological conditions and controls were appropriate, with no discrepancies noted. Various maintenance in contaminated and high radiation areas was observed during the maintenance outage. The inspector witnessed use of the optical disk surrogate tour in preparation for drywell and reactor building inspection activities. Use of the surrogate tour by inspectors and maintenance personnel was effective in reducing personnel radiation exposure. Radiological briefings and support during maintenance were outstandin .1 Unanticipated liigh Itudiation I etels in Radwaste Deminerallier Room On October 29,1992, two personnel reported that their self indicating dosimetry, set to alarm at a dose rate of 50 millirem per hour (mR/hr), had alarmed when entering the radwaste demineralizer room. This was unexpected since previous radiological surveys had indicated that typical dose rates in this area were 10 to 20 mR/hr. The room was posted as a high radiation area. A radiological protection (RP) technician immediately performed a survey and confirmed that the dose rates in the radwaste demineralizer room were signincantly higher than norma The area dose rate at the room entrance step-off-pad was 1200 mR/hr while the highest on contact dose was 70 R/hr at the radwaste demineralizer. A door guard was immediately stationed and the room was then controlled as a locked high radiation area. Dosimetry was proecssed for the two workers who had initially reported the high dose rate condition. Their individual exposures (1 mR and 3 mR respectively) were well below 10 CFR 20 regulatory limits. The Radiological Section hianager was informed, and a door kicking device was installed. All personnel involved responded promptly and took appropriate corrective action _ . _ _ _ __ ._

_ _ _ _ __ _ a

.

A critique was initiated to determine the cause of the higher than expected radiation levels in the radwaste demineralizer room. The inspector noted licensee discussion to be detailed and pursuant to a logical conclusion. The critique identiGed the most likely cause of the high radiation levels to be the direct processing of unnitered reactor water by the radwaste demineraliter. The reactor water cleanup (RWCU) system deminerallier is normally used to Olter reactor water during both power and shutdown conditions. Continued direct treatment of reactor water through the radwaste demineralizer (during a period when the RWCU demineraliter and the condensate deminerallrcr were unavailable) resulted in higher than normal radiation levels in the radwaste demineraliter roo The inspector reviewed the RWCU system maintenance work package and noted that the potential of higher radiation levels in the radwaste demineralizer room had not been considered during the RP review. It appeared to the inspector that the RP review should have identined ;

this potential condition. Review of maintenance records, operations logs, and RP survey records

__

for the current operating cycle identined no previous occurrence of this same conditio *

Recommendations to preclude recurrence included revision of operating procedure 2.2.19,

" Residual Heat Removal System", to inform RP and radwaste personnel prior to transferring reactor water to radwaste. In addition, Radiological Section standing order 92-30 was issued to specify radiological posting and control of the area when the radwaste deminerallrcr is used as a scactor water letdown path. The inspector reviewed the revision to procedure 2.2.19 and . ^

standing order 92 30 and detennined that the new instructions were clear, concise, and correctly addressed the causal condition. The critique was effective and corrective actions were promptly implemente .0 MAINTENANCE AND SURVEILI,ANC'E (37828,61726,62703,93702) Motor Operated Vahe Maintenance The licensee undertook an aggressive motor operated valve (MOV) workload during the mideycle outage encompassing maintenance and surveillance on 52 MOVs. Inspections and 1 corrective maintenance to resolve generic issues were performed in addition to several MOV t actuator overhaul Eleven MOV torque switches were inspected for the presence of Ober spacers as initially discussed in 1.imitorque Corporation 10 CFR 21 noti 0 cation dated September 29,1989. Two of these torque switches contained fiber spacers and were replaced with torque switches of a newer design containing a metallic contact bridge in place of the Ober spacers, The last MOV remaining to be inspected for this issue (MOV 406011) has been scheduled for inspection'during planned valve maintenance in_ january 1993. NRC Inspection Report 50 293/92 80 previously concluded that the licensee had not taken action to resolve the fiber spacer issue in a timely _

manner. Additionallicensee inspection activities and resultant torque switch replacements during ,

the March 1992 maintenance outage and during the current outage have appropriately addressed this issu ,

__

.

.

Thirty additional inspections for generic concerns were completed, addressing a 1.imitorque 10 CFit 21 notincation dated December 11,1990 regarding undersited torque switch roll pins and NRC Notice 88 84 concerning dc0cient motor pinion keys. The inspector concluded that the MOV corrective maintenance program is appropriately addressing generie industry concerns, liight h10Vs were overhauled as part of the h10V preventive maintenance program. The inspector reviewed completed work packages for maintenance performed on the outboard containment spray isolation valve (hiOV 1001 2311) and the reactor core isolation cooling (RCIC) injection valve (MOV 1301 48). During overhaul of the hiOV 1001-2311 actuator, the spring pack was replaced, a torque switch limiter plate was installed, and minor physical discrepancies were properly corrected in accordance with approved procedures, post ~

maintenance diagnostic testing was tuformed in accordance with procedure 3.M.3-24.12,

Vol!!S 100 Operating l'rocedmf with satisfactory results. Preventive maintenance was performed on the MOV 130148 limit switch assembly and the motor pinion key was replaced -

(reference: Notice 88-84). One isolated case of incorrect maintenance status-resulted in-operation of MOV 1301-48 prior to adjustment and testing of the limit switch, which in turn caused motor damag The motor was promptly replaced and lmst-work testing was satisfactorily accomplished. The cause of the miscommunication was identified and appropriate cmrective actions were initiated. The inspector reviewed substitution equivalency evaluations for the motor and the motor pinion key (SSii-530 and SSii-548 respectively) and determined that the bases for equivalency were appropriate. Overall coordination of MOV maintenance was excellent, supporting completion of a significant amount of corrective and preventive maintenance during the outag The inspector observed an MOV program team critique of outage maintenance activities. The critique was detailed and prmiuced gmxt recommendations to further enhance work prmluctivit Reconnuendations included prestaging of spare overhauled MOV actuators for in kind

-

substitution; use oflocal vice remote MOV control during post-work testing; implementation of a consolidated MOV bolting specification; climination of redundant operations and maintenance operability testing; and, MOV technician training enhancements. The MOV Program Manager plans to further develop and present recommendations to licensee management for disposition and implementation. During the outage, the licensee implemented an aggressive schedule and completed a noteworthy amount of MOV maintenance, accomplishing improved valve reliabilit A comprehensive preventive maintenance scheudle remained under development due to pending revisions to the MOV program scop .2 Reactor Water Cleaoop System leakage As documented in NRC inspection reports 50-293/92-14 and 92-16, a small coolant leak on the shell of regenerative heat exchanger 11-20811 in the reactor water cleanup (ItWCU) system has been monitored by the licensee since its discovery in June 1992. Continued safe operation of the facility was not adversely affected by the identified leakage conditions and repair operations were scheduled to be conducted during the outag I

,..

.

..

During this inspection period, which encompassed outage maintenance and repair activities, the inspector reviewed the maintenance request, governing the repair of RWCU heat exchanger E-2088. While this second stage component to the RWCU regenerative heat exchanger is a nonsafety-related piece of equipment, it was originally procured as an ASME Section 111, Class C vessel and is a code stamped item. Therefore, ASME Section XI repair / replacement rules governed the repair plans without need for additional section ill code certincation. A code reconciliation to a later version of the ASME Section lit code (i.e.,1980 edition through winter 1980 addenda) was performed and the 1980 edition of the ASME Section IX code was invoked for welding activitie The inspector reviewed the maintenance work plans, controlling heat exchanger disassembly and diaphragm inspection, and the governing the leaking diaphragm weld repair. One MWP that discussed an option to replace, rather than repair, the existing diaphragm was not used. As was discussed in IR 50-293/92-14, a defect in the stainless steel diaphragm, serving as the non load carrying pressure boundary, was suspected and subsequently con 0rmed, to be the source of the identified leakage. The inspector also reviewed the MR log sheet, the weld repair data sheet and traveler no. 3432.03. These records, along with the ASMB XI repair forms, were evaluated for consistency with the technical criteria delineated in the ASME Code and ESR Response Memorandum, ERM 89 1171 (Revision 1) documenting the engineering repair recommendation .

The inspector questioned some informational data (e.g., diaphragm material type) on certain of the supporting documents to the weld repair records and concluded that repair activities were adequately controlled. The source of the leak was found to be a pinhole defect in the stainless steel diaphragm, which was excavated and liquid penetrant tested (LPT) in accordance with ERM and code provisions. The inspector examined the welding procedure speci0 cation procedures and verified acceptable LPT of the Onal weld, along with QC coverage of the required NDE and specific welding variables, e.g., preheat and interpass temperatures. Since ERM 891171 (Revision 1) was written to address any future repairs to any of the three stages to regenerative heat exchanger E 208, the inspector also examined the GE specification sheet in the Heat Transfer Products vender manual, and MR document packages to confirm appropriate consideration of the currently installed diaphragm material type. The inspector noted that ASME/ ASTM SA 240, TP-304 stainless steel was originally provided as heat exchanger diaphragm material and subsequently changed to TP-304L low carbon material to address integranular stress corrosion cracking (IGSCC) concerns. As documented in the governing ERM, the option for replacement of the diaphragm, if required for future heat exchanger repairs, involves a material substitution to SA 240, TP-3161sstainless steel, representing the best option to prevent corrosion pitting and the propagation of IGSCC defect The inspector assessed the engineering direction, sequence of work and QC . coverage, and overall control of heat exchanger E-208B heat exchanger repair operations and identified .no unresolved safety issues. Prior to startup from the outage, the RWCU system was placed back in service with no evidence of a continuing leakage problem. The inspector has no additional questions regarding either the acceptability of the repair work or the documentation package completenes ,

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

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

i

-

)

l

.

8 , Mnintenance and Modification Field Work i The inspector conducted several plant inspection-tours during the outage, checking the status of plant equipment and the conduct of maintenance and modification activities in progress. The

'

following components, work and documentation were spot-checked by the inspector to ensure adequate planning and control of the observed activities:  !

-- Machine tooling of the collector ring on the exciter end of the "11" recirculation pump motor-generator (M-G) set (X 204B) in accordance with maintenance work plan, MWP 19201023-1,

-- Installation of concrete expansion anchors and conduit supports as part of the plant design change, PDC 9159A, for the replacement of transformer X56, providing 120V vital AC power to distribution panel Y4, in accordance with maintenance request, MR 19104143,

-

QC initiation of a problem report, PR 92-9189, and the nuclear engineering department '

.

(NED) conduct of an operability evaluation for certain reactor building closed cooling water (RBCCW) pipe supports identified with structural configuration deficiencies (reference work request tags 027809 thru 027812), and

--

Operations control (tagout "192-002 015) of the 'll" loop recirculation M G set h>ckout relay testing in accordance with procedure 3.M.3 3 .

The inspector interviewed craftsmen, operators, field engineers and- QC personnel, as appropriate, to evaluate the acceptability ofimplementation of the above work activities. Field documentation and records were spot-checked and procedures examined, as necessary, to confirm applicability to the relevant quality criteria. Where work remained to be completed, the inspector noted the provision for the conduct of the appropriate post-work testin Continuing open items were documented (e.g., PR 92.9189) and evaluated for operability. The inspectoi identified no unresolved safety concerns and had no additional question regarding the conduct or status of the observed field work. The MR/MWP documents examined appeared to be well organized, written cicarly and capable of providing both effective direction to work crews and acceptable recorded evidence of the quality work performe ,

.

k

---

%,- w, - - -,, w , y ,,m. %,_, , , e.,,., ,m _

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

'

!

9 EltGl:NCY PREPAREDNESS (40500)

On November 19, 1992, the licensee conducted an annual test of the Prompt Alert and Notification System (PANS) for the ten mile emergency planning zone (llPZ) surrounding Pilgrim Station. The PANS system consists of remote activation consoles located at the five EPZ town fire department dispatch centers and 112 notification sirens, The test was conducted to verify the PANS capability to notify occupants of the liPZ in the event of an emergenc Procedure !!P-AD 417, " Annual Siren Test Program", requires at least 90 percent of the PANS sirens (or 101 sirens) to operate properly for the test to be successful, if this acceptance criterion is not met, the system has failed and the licensee would be required to notify the NRC in accordance with 10 CliR 50.72. The inspector questioned the clarity of the wording of test acceptance criteiion specified in lip-AD-417, noting that it could result in unnecessary town, state and NRC notifications. The licenwe initiated appropriate actions to clarify the test acceptance criterion. The inspector reviewed siren feedback system records and observer log sheets from the test and noted that all but one of the 112 sirens had properly sounded. An obwrver in the vicinity of the single failed siren reported that the warning tone could be heard, llowever, the peak decibel level recorded by the automated siren feedback system was outside of normal parameters. Therefore, the licensee conservatively categorized that single siren as a failure and initiated corrective action to repair the siren. The iispector concluded that the test was successful and that the licensee personnel demonstrated a high level of knowledge regarding the design of PANS .md test performanc .0 SECURITY (71707)

Selected aspects of plant physical security were reviewed during regular and backshift hours to verify that controls were in accordance with the security plan and approved procedures. This

-

review included the following security measures; security force staffing, vital and protected areas barrier integrity, maintenance of isolation zones, behavioral observation, and impicmentation of access control including access authorization and badge issue, searches of personnel, packages and vehicles and escorting of visitors. Security force personnel continued to perform their duties in an alert manne .1 Stodification Compensatory Steasures During inspection of certain piping replacement activities, the inspector evaluated the adequacy of compensatory measures where security areas had been breached by the construction wor Augmented access controls were in place, and the inspector confirmed that compensatory controls would continue to be implemented until a more permanent measure was installed to prevent area ingres The inspector questioned whether the suspension of full access compensatory measures met the intent of NRC Regulatory Guide (RG) 5.65 for barriers at water sources. The inspector discussed both the regulatory guidance and the planned licensee security controls for accessible

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

.

.

piping with the PNPS Security Section Manager and Region I security specialists. It was noted that at the time of this inspction, the system was not required to be operable and was in fact out of service to implement the planned piping replacement modifications, llowever, after return to service, a portion of the piping would remain accessible from outside the area barrier Therefore, the licensee committed not only to reinstitute acceptable compensatory access controls, but also to continue implementing those security measures necessary to safeguard the piping appropriatel The inspector determined that the overall licensee security controls and compeasatory measures for the piping replacement activities provided adequate safeguards protection. IlECo's licensee responsiveness to NRC concerns was timely and well directed. The inspector verified that the licensee has plans for continued security coverage of the piping replacement activities, in accord with the guidance of RG 5.65 and the provisions of the Pilgrim Security Plan, until the modification is completed during the next refueling outage (RFO 9). Illness For Duty Prognun Review During the week of November 2 7, 1992, a regional-based security inspecter conducted a selective review of the licensees fitness-for-duty (FFD) program implementation. The review consisted of an examination of the licensee collection facility, collection and chain of custody procedures, and records. Additionally, the inspector interviewed licensee medical staff members concerning specimen collection and chain of custody procedures. The inspector determined that the individuals were very knowledgeable of their specine duties and responsibilities. Based on the foregoing review, the inspector determined the program was being effectively implemented and was in accordance with the requirements of 10 CFR Part 2 .0 SAFETY ASSESSMENT AND QUALITY VERIFICATION Assessment of Midcycle Outnge Activity Controls Prior to the midcycle outage, the licensee established a system " window" approach to maimain maximum decay heat removal capability, reactor coolant system inventory control, electrical distribution availability, and primary and secondary containment control during outage related maintenance and testing. The two major phases or windows to the outage were "A" train and

"11" train system outages. The window approach during the last refueling outage was very effective in providing maximum con 0guration Dexibility of available Guid and electrical system Additionally, the process clearly established system configurations which met or exceeded technical specification requirement The licensee outage safety review team, chaired by the onsite review committee chairman, independently reviewed the outage schedule to assess safety perspective and license complianc The team also reviewed the midcycle outage compensatory measures procedure (TP 92-067)

established to provide operations personnel with an overview of the outage schedule, system unavailability, established compensatory measures, and mitigation capabilities. The team

!

.

I i

-

..

identified ten enhancements to the outage operations. Significant team contributions included maximization of reactor vessel inventory to improve loss of shutdown cooling mitigation capability, maximizing power source availability, several contingencies to address potential i system misalignments due to valve mispositioning, and awareness of contingencies during electrical bus maintenance outage ;

The lleensee also established an emergent issues team (filT). This team coordinated the response to outage issues on a real time basis in order to minimize overall outage management impac To a lesser extent, the tilt projected potential concerns and contingencies to ongoing outage activities. The approach to reactor startup was well controlled. Completion of post mod 10 cation ,

and maintenance testing was verified. Train and system restoration and valve lineups were accomplished in an orderly fashion. Additionally, the ORC reviewed the status of the reactor vessel water level instrumentation prior to startup. Overall, licensec control of outage activities with respect to plant safety was concluded to be excellen .2 Verification of Pinnt Records (Tl 25 5/l15, RI-TI 92-01)

On April T.3, the NRC issued Information Notice IN 92 30 " Falsification of Plant Records,"

that documented instances of falsified plant logs at several nuclear power plants. The inspector documented initial licensee response to this issue in section 7.3 of NRC Inspection Report 50-293/92-0 The NRC's resident inspector staff have accompanied plant operators on selected portions of plant tours, including a complete deep backshift accompaniment of the turbine building round Throughout all direct inspection observations, operators were noted to be knowledgeable of tour procedure (procedure 2.1.16) requirements, inoperable equipment was properly identified, changing system parameters were appropriately documented, and general housekeeping such as component wipe-downs and loose debris collection was good. Additionally, the operators were

_

noted to have made excellent use of the closed circuit television (CCTV) tours of the condenser bay, reactor water clean up system (including heat exchanger leak assessment), and the turbine ,

deck. The value of the CCTV tour with respect to comprehensive area inspections in low radiological dose rate environments was clearly eviden Throughout this period of time, IlliCo Quality Assurance Department (QAD) continued to -

conduct surveillance of operator tour A QAD surveillance of operator tours from Sep!cmber 13 26 was conducted as documented by QAD surveillance number 92-1.1-12. The surveillance reviewed operator entries into the reactor building, diesel generator rooms, vital MG set rooni, "A" and "II" switchgear rooms, salt service water pump rooms, main stack, and several other plant areas as described by tour log data entry and by plant security computer printou Two inconsistencies were identified during the surveillance by issuance of deficiency report (DR)

No. 2004. The first inconsistency involved operator documentation of the main stack tour. One-logged entry into the main stack facility was not recorded by the security computer. The a

V

.

• 12 security computer recorded the operator exiting the main gate, as would be reouired to accomplish the remote location tour at a time consistent with travel to the main stack. At the time in question, maintenance was being performed at the stack and the facility roll-up door was open. The operator stated that entry to the facility was accomplished directly through the open door, and that the security card reader was inadvertently bypassed. The QAD independently verified that the door was open and that operator exit and return to the main gate were consistent with the times in question. Therefore, the first inconsistency documented in DR 2004 was determined to have an acceptable explanatio The second irconsistency involved a single entry into a switchgear room during a September 21, 1992, turbine building tour. The security computer <did not record the operator's entry or exit from the area during the time that the tour was conducted. The remainder of the tour was verified to be consistent with security computer records. Operations Section management responded to the deficiency by conducting a full review of the subject operator's tours for calendar year 1992. The review indicated 100% compliance with' log entry and security computer records, except for the one noted discrepancy. Additionally, the individual in question (who holds a reactor operator license) maintains having completed entry into the switchgear room during the tour in question. Further review indicated that security controls for entry to the switchgear _ room h?d been verified to be functioning properly, consistent with normal periodicities. The switchgear room check did not involve any surveillances required by Technical Specification The QAD surveillance also identified certain weaknesses in the Operations Department self-monitoring program of operator tour entries. Specifically, the monthly nuclear watch engineer (NWE) assessment of tours for July, August and October were properly completed, but the September 1992 NWE assessment had not been performed. The quarterly Chief Operating Engineer assessment for the same period had also not been performed. Additionally, the monitoring of operator tours and turnovers were being performed weekly, vice three times per week, as instructed by Prc .edure 1.2.4, " Operations Performance Assessment Program." The QAD issued a second deficiency report (DR No. 2005) to document these self monitoring weaknesse Independent assessment of record verification by the QAD has been effective. Initial response to NRC IN 92-30 was comprehensive and confirmed that operators are properly completing -

required plant tours. The surveillance process has maintained objective criteria such that any -

noted observations, weaknesses and deficiencies have been appropriately communicated to the-Operations Section.

l

- ,

.

..  !

13 ENGINEERING AND TECilNICAL SUPPORT (71707) Reactor Vessel Water lael Instrumentation Spiking R l

8.1.1 Itackground As documented in NRC inspection report 50-293/92-04, the licensee has experienced reactor vessel water level instrumentation " spiking" during reactor depressurization following plant shutdowns. Typically, the level instrumentation spiking has been observed to begin at reactor pressures below 470 psig' during depressuriwions. The spiking initially occurs on the level instrumentation associated with the reactor vessel "B" re r cv leg. The spikes are typically of 20-30 seconds in duration, four to six inches in ampliS de A.J are similar to a square wave recorder trace. As reactor pressure decreases, the amplitude of the spiking increases. At reactor

'

pressures below approximately 60 psig the recorder trace of the spiking becomes less regula Instrumentation associated with the "A" reference leg initially exhibits the spiking phenomenon at approximately 65 psig and is typically one to two inches in amplitude. Maximum spiking observed at very low pressures (i.e. less than 60 psig) has been approximately 22 inches on the-

"II" reference leg instrumentation and 17 inches on the "A" reference leg instrumentatio Licensee corrective actions for the spiking phenomenon have been primarily focused on the improved thermodynamic performance of the condenting chambers and- steam drain liries associated with the level instrumentation system, llowever, throughout the evaluatio_n of the spiking phenomenon, the presence and accumulation of noncondensible gases within the reactor instrumentation system has been recognized as a potential contributing factor. Temporary .

temperature instrumentation was installed on the " A" and "B" reference leg condensing chambers and associated steam drain lines in order to monitor and trend condensing chamber performanc The instrumentation also moniars the potential effects of the buildup of noncondensible gases within the reactor instrumentation system during power operations, as _well as during ensuing plant shutdowns. Additionally, the licensee management team (or " Issue Team") assembled to-investigate this issue remained active and contracted _with an _ instrumentation engineering -

consultant with expertise specific to boiling water reactor (BWR) level instrumentation system design and operatio .1.2 Generic Applicability Subsequent to the April 12,1992 station startup, generic concern for potential inaccuracies in -

boiling water reactor vessel water level indication during and after rapid depressurization events was identified. Specifically, the concern addressed the evolution of noncondensible. gases, accumulated in the reference legs, during reactor depressurization in sufficient volume to cause errors in indicated reactor vessel water level. This concern is d(x umented in NRC Information Notice 92-54, " Level Instrumentation inaccuracies Caused by Rapid Depressurization," dated July 24,1992 and-NRC Generic Letter 92-04, " Resolution.of the issues Related to Reactor Vessel Water level Instrumentation in IlWRs Pursuant to 10 CFR 50.54(f) " dated August 19, 199 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -_ _ .__ _

\ .

t

Continuing engineering analysis by BECo (and their consultant) preliminarily concluded that the potential exists for noncondensible gases to accumulate in the reference legs. Independent NRC review of the available Pilgrim-specific and generic data and analyses, as documented in inspection report 50-293/92-17, concluded that noncondensible gas accumulation in the reference legs signincantly contributed to the recent ratetor vessel water level instrument spiking event The NRC also determined that current station data strongly suggest a potential problem may still exist, especially in the "B" reference le The noncondensible gas accumulation phenomenon appears to be dependent upon three basic contributing effects including: 1) the presence of noncondensible gas in the condensing chambers; 2) a mechanism for transporting the gas into the reference leg, such as small leaks in reference leg Ottings and components; and 3) an instrumentation piping geometry that supports gas accumulation (e.g., horizontal sections). Licensee consultant and generic evaluations indicate that the elimination of any one of these effects can ameliorate the spiking, in response to this information, BECo took action to identify all external leakage paths on reference leg piping located outside the drywell. An approved developer material, that reveals leaking component interfaces, was applied to approximately 400 joints on the reference legs and instrumentation racks. Joints that exhibited leakage were bagged and the water leaks were quantified. Results indicated the "B" reference leg external leakage to be approximately 100 milliliters per day and " A" reference leg external leakage to be approximately 0.14 milliliters per day. A temporary procedure to be conducted during the midcycle outage was developed to quantify leakage internal to the instrumentation system (i.e., past the individual instrument bypass valves). Additionally, BECo installed ultrasonic equipment on the reference legs with the intent to detect potential gas bubble travel during reactor depressurization after plant shutdown for entry into the recent midcycle outag .1.3 Midcycle Outage Reactor Shutdown and Depressurization On October 23-24,1992, the licensee performed a routine reactor shutdown and depressurization for the purposes of conducting a scheduled 30 day surveillance testing and maintenance outag The reactor had been at power operation for 190 days prior to the shutdown. Reactor vessel water level instrumentation was monitored closely during reactor depressurization. At 10:45 am on October 24, the reactor mode select switch was placed in the SHUTDOWN position. rcom approximately 10:58 am to 11:14 am the reactor was depressurized from 750 psig to 350 psig with no water level instrumentation spiking observed. However, at approximately 350 psig, small spikes of two to four inches in amplitude were observed on the "B" reference leg level instrumentation. Similar amplitude spikes continued to occur over the next five hours as reactor pressure was decreased to 65 psig. The amplitude of the spikes increased to 8-17 inches as reactor pressure was reduced to 30 psig over the next two hour _- - _ - _ _ _ _ - _ _ - _ - _ _ _ - _ _ _ _ _ _ - _ ._ . _ _ _ _ __ _ _ _ _ __ _ _ _ _ _ _ _ - _

_ .

,

At 10:28 pm on October 24, with the reactor essentially depressurized at 2 psig and reactor coolant temperature at 165 degrees F, reactor vessel water level as indicated on "B" reference-leg instrumentation spiked 29 inches (from +21 inches to +50 inches), resulting in an automatic primary containment isolation system Group 1 isolation signal._ Spiking of two inch amplitude was first observed on "A" reference leg level instrumentation at approximately 63 psig reactor-pressure. Several spikes of 17 inches were observed on " A" reference leg level instrumentation at less than 5 psig reactor pressur The reactor vessel water level instrumentation spiking observed during the October 24 depressurization was similar to that experienced during recent reactor shutdowns. Although the spiking began at lower pressures and was initially _ of lesser amplitude, the signature of the spiking recorder traces was essentially identical to the previous occurrences. Past corrective actions to improve condensate chamber and steam drain line performance by- addressing thermodynamic performance appeared to have been minimally effective. Additionally, the ultrasonic equipment confirmed the travel of gas bubbles in the "B" reference leg. Following completion of the reactor cooldown and fmal analysis of level instrumentation data, the licensee concluded that the primary cause of level spiking was noncondensible gases comino out of L solution during reactor depressurizatio .1.4 Corrective Actions As was discussed previously, the volume of noncondensible gases present within the reference legs is significantly innuenced by reference leg configuration and by the presence of very small leaks in reference leg fittings and components. These relatively minor reference leg fitting leaks provide a slow and persistent flow which causes the gases to migrate down the reference leg BECo corrective actions during the midcycle outage were directed toward minimization of both

_

" external" and " internal" reference leg leakage paths.- External leakage paths identified during plant operations were repaired primarily by tightening fitting interface Internal leakage (from the reference leg past instrument bypass valves to the variable leg) was quantified by performance of a leak rate test. Temporary procedure TP 92-047 sequentially established instrument rack electrical and mechanical isolations and applied a' demineralized water pressure test source of 20 psig to the low side head tank of a selected instrument on the-instrument racks (RK 2205,2206) associated with the "A" and "B" reference legs. ' Testing of

.

the "B" reference leg instrument racks was initiated on November 9. Test pressure and head tank level stabilization were established and test parameters were monitored for approximately-8 hours, with no quantifiable internal leakage observe Testing of tne "A" reference leg instrumentation racks was initiated on November 15. Test conditions were established and test parameters were monitored for approximately 15.5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> The " A" reference leg high side isolation s alve (2-HO-126A), which established a test boundary, ,

appeared to be not fully seated. The inability to fully isolate this valve created _a leakage path j

'

'' '

- "

--.m_.___.___._____-________ _ . _ - _ _ . . _ _ . _ _ _ _

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

.

.

for the test demineralized water source from the reference leg to the reactor vessel. This leakage was not directly relevant to the test objectives but, if quantifiable, could be used to estimate bypass leakage. Initial "A" reference leg instrument rack leakage was quantified as 21.6 ml/ day, which included instrument bypass valve leakage as well as leakage past the 2 HO-126A valve to the reactor vessel. In order to more accurately quantify bypass valve leakage, the licensee isolated each level instrument and calculated any leakage rate change. With all instruments isolated, leakage rate was calculated to be approximately 14 ml/ day. Therefore, bypass valve leakage on " A" reference leg instrument racks was assumed to be approximately -

7 ml/ da llowing subsequent plant startup and pressurization, the licensee intends to inspect for the presence of external leaks and quantify any observed leakag .1.5 issue Status On November 20, the DECO reactor vessel water level instrumentation spiking Issue Management Team presented a status report to the onsite review committee (ORC). The report (attached to this inspection report) summarized data from the October 24 shutdown. The Issue-

,

Team found that instrumentation response during reactor depressurization was consistent with l recent shutdowns, and that the characteristic spiking signature was repeatable. Spiking observed ~

during the October 24, 1992 shutdown was bounded by previous operability analyses which assumed the presence of noncondensible gases in the reference legs. Therefore, the licensee l concluded that the level instrumentation remained operable throughout the October 24th shutdow The status report addressed corrective actions completed during the midcycle outage including; repair of reference leg external leakage paths, quantification ofinternal instrument bypass valve leakage, and reinstallation of steam drain line insulation. The insulation had been previously removed in order to improve condensing chamber performance by increasing heat transfer across

'

the steam drain line and condensing chamber on the: "B" reference leg. However,

-

instrumentation performance did not demonstrate any significant improvement during the October 24 shutdown; therefore, the insulation was reinstalle The status report provided a detailed review of Pilgrim specific experience relative to the generic concern for level instrumentation inaccuracies due to noncondensible gases as described in NRC Generic Letter 92-04. The BECo issue Team concluded that the level instrumentation response during recent Pilgrim shutdowns was consistent with the noncondensible gas theories presented to the NRC staff by the Regulatory Response Group of the Boiling Water Reactor Owner's Group (BWROG), and was similarly consistent with the theories developed by the licensee- .

contracted specialist with expertise in this discipline. The Issue Team also concluded that the instrumentation spiking observed at Pilgrim would not affect either the limiting FS AR transient and accident analysis or the operability evaluations and conclusions of the plant-specific safety assessment (as well as the BWROG generic safety assessment) in response to NRC Generic Letter 924) .

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

17 The ORC discussion of the level instrumentation concern with the issue Team was observed by the NRC resident staff to be well focused. Extensive technical detail from NRC Generie Letter 92-04 was requested by the ORC Chairman, and discussion regarding NRC staff acceptance of interim operation with the noncondensible gas phenomenon as described in Generic Letter 92-04, such that the committee was enabled to independently establish generic applicability. The committee also reviewed level instrumentation response to design bases transients and accidents with the issue Team. Following additional description of Pilgrim experience, and plant-specine -

safety assessment, the ORC independently concluded that the safety function of the level instrumentation would be fulfilled and that the issue Team continues to effectively manage ,

station response to the phenomenon. Following ORC recommendation, the Plant Manager approved the issue Team reactor water level spiking status repor NRC inspection, evaluation and assessment of license performance with respect to the spiking occurrences, to date, have identified no violations of Pilgrim license conditions. . The NRC staff .

also independently reviewed the bases for BECo's operability determination, and agreed with its conclusions. The NRC has determined that the plant-specific assessmeat for Pilgrim has acceptably addressed regulatory requirements (and the points raised in Generic Letter 92-04),

- and has verined the continued safe operation of the plant while the spiking phenomena'is-investigated further as a generic issu .2 Salt Service Water Piping Replacement As documented in inspection reports 50-293/92-08,92-10, and 92-14, salt service water (SSW)

piping replacement activities commenced in May 1992 and have continued in several stages to the present. During the outage, the licensee installed the first pieces of titanium piping as."B" train penetration spools in the intake structure and auxiliary bay walls. As governed by plant:

design change (PDC)91-10C, this phase of work involved removing existing rubber-lined, carbon steel sections of piping and replacing each penetration spool with titanium pipe, an expansion joint member and a new rubber-lined, carbon steel pipe piece to reconnect to the existing SSW supply headers. The inspector witnessed the modification activities, noting that the "B" SSW train was returned to serme during the outage with the titanium penetration spools

_

installed and that the corresponding "A" train work is scheduled during RFO 9. During the period of time between the outage and RFO 9, new titanium piping 'will be installed-in excavations in proximity to the existing SSW headers in accordance with controls established by

PDC 91-10D. The final tie-in of the titanium ' piping to existing SSW piping in both the intake-

! structure and auxiliary bays is planned for completion during RFO 9.'

' The inspector reviewed PDC 91-10C-02 and related portions of PDC 91-10D. The inspector.

L also reviewed the current revisions of the original piping specifications, M-300 and M-301, and l the titanium piping specification, M-100E. along with the commercial grade. item (CGI)

L engineering evaluation No.- 596 for the titanium pipe and fittings. As required by specifications,-

[ - - - the inspector verified that the installed titanium piping penetration spools had been procured to American Society for Testing of Materials (ASTM) B337-83 standards. However, both the FSAR change associated with safety evaluation 2693 and the acceptance criteria for pipe l

- , - - . -

. . - - .- . - -

'

,

,

chemical properties listed in CGI 596 indicated that the 1987 edition of the ASTM standard was the correct reference. The inspector discussed this discrepancy with licensee compliance personnel who initiated Problem Report 92-0536 to resolve the noted document conflicts From a technical standpoint, the titanium spool pieces were fabricated and supplied to the correct design standards. The inspector reviewed the "A" SSW supply header drawing, SKM-91-10C-001C, for the intake structure spool / expansion joint assemblics and compared the listed material-requirements with the referenced specifications. No hardware deficiencies were identified and the inspector has no further questions regarding component fabrication detail The inspector also evaluated the ongoing SSW pipe replacement activities with respect to PDC 91-10C details, governing hydrostatic pressure testing to ANSI B31.1 and ASME Section XI code requirements, protection of the titanium piping from galvanic corrosion, and verification that the unlined carbon steel spool pieces remaining in service meet code minimum wall requirements. With regard to the latter inspection point, the inspector reviewed the BECo response to NRC Generic Letter (GL) 8913; NED memorandum 91-145, revision 1, outlining routine SSW piping inspection plans; and BECo quality control inspection report 92-60, documenting the results of the ultrasonic thickness examinations of the SSW pipe spools installed without rubber lining. The inspector verified that actions to address the concerns documented in GL 89-13 were consistent with licensee commitments, program controls and the phased sequence of SSW piping replacement activitie Additionally, the inspector witnessed the re-installation of a hanger, H29-1-12SG, for the "B" SSW train piping. The inspector confirmed adequate work controls were specified in the main enance work package, MWP 19200471-1, and observed craft compliance with engineering criteria (e.g., torque requirements) and adequate QC coverage of the job. Ongoing SSW "B" train pipe replacement and restoration activities were evaluated for impact on the "A" train SSW system and components required to be operable and operating. The inspector identified no adverse consequences on the function of the SSW ,ystem to provide decay heat transfer to the ultimate heat sink during the outag The overall SSW piping replacement project, to date, has been well planned and conservatively executed. The inspector reviewed a sample of nonconformance reports, written against vendor materials supplied to support PDC 91-10 work.. Adequate QC inspection of safety-related material and appropriate engineering evaluation of the sampled NCRs were found to have been l implemented. -The inspector interviewed craft, QC, engineering, and project management

_

i personnel involved with PDC 91-10 activities and identified no unresolved safety concerns. At the conclusion of this inspection, ongoing excavation work governed by PDC 91-10D continued.

l Future NRC inspections by regional and resident personnel will monitor the progress.of this

'

major modification, o

i

!

- - .-

.

.

19 NRC MANAGEMENT MEETINGS AND OTIIER ACTIVITIES (30702)

91 Routine Meetings At periodic intervals during this inspection, meetings were held with' plant management to discuss licensee activities and areas of concern to the inspectors. At the conclusion of the reporting period, the resident inspector staff conducted an exit meeting on December 11 with IlliCo management, summarizing the preliminary inspection findings for this report period. No proprietary information was , identified as being included in the repor .2 Other NRC Activities On November 2-6, two NRC Region I radiological specialists conducted an inspection to evaluate licensee radiological controls practices during outage conditions. Inspection results will be documented in Inspection Report 50-293/92-2 On November 2-6, an NRC Region I security specialist conducted a programmatic security inspection. Inspection resul:s are documented in Inspection Report 50-293/92-2 On November 5, NRC resident inspectors attended a meeting of the Plymouth Nuclear Matters Committee. The committee discussed issues which had been raised by the Citizen's Radiological -

Monitoring Network, Inc. (CRMN) in a letter to the NRC regarding events which occurre during the October 24 Pilgrim shutdown. The NRC responded to the CRMN letter on November 2 On November 16-20, NRC Region I operator licensing specialists administemd initial license examinations and requalification examinations. Results of the examinations will be documented in Inspection Report 50-293/92-2 On November 16-20, an NRC Region 1 fire protection specialist conducted an inspection of the Pilgrim fire protection program. Inspection results will be documented in Inspec; ion Report 50-293/92-2 On November 19, two NRC Headquarters officials accompanied the NRC resident inspector on a safety inspection of the drywell and selected portions of the reactor buildin I

.

.

^,.

Office Memorandum Boston Edison Company gg

m To: E.S. Kraf t, J fro /:S.D.budson Record sypo A4'08

.

Date: November 20, 1992 Dept. Do ESE092-167 Hon-Safety Related Subject: REACTOR WATER LEVEL SPIKING STATUS rep 0RT Distritiulion:~~

R.A. Andarson E.T. Boulette W.S. Clancy J. Hellafeuillo W.C. Rothert R.V. Fairbank H.V. Oholm T. Sullivan D. Tarantino J. Rogers-RT S. Dasgupta-8T R. Cannon T. Trepanier R. Kolley-BT K. Kampschnoider E. Almolda P.D. Smith L. Francis Approval For Reinase Outside Plant Dep F.S. Kraft, J BACKGROUND:

~

Dur1ng the shutdown and cooldown on 10/24/92, reactor water level was continuously monitored with the EPIC Computer. Close communication was maintained with the Control Room. Spiking was observed to begin at approximately 350# on the "B" ECCS instruments. The characteristic signature which had boon observed in the past was present during depressurization. The overall performance of the water level instrumentation was similar to past performanc '

.A Group 1 Isolation occurred at approximately 160"f dua to a maximum recorded spike of 29 inches (limited by instrument range). Water level was 421 inches at +ho time of this spike. All valves responded as designed due to the falso high' level signa Spiking began at 63# on the "A" Side instrument The "A" Side also performed as expecte The maximum spike observed on the "A" Sido was 17. inche OPE MDILITY DETERMINATION F0110 WING OCTOBER 24. 1997 SHUTDOWN:

Reactor water level instrumentation remained operable during the shutdown and

.cooldown on 10/24/92. An Operability Evaluation was. performed by Engineering and Plant personnel in September 1992 which assumed the existence of non-condensible gas in the reference legs. This evaluation recommended operability criteria- for the reactor water level instrumentation if notching were to occur during a plant shutdown. The ORC reviewed and concurred with the evaluation-and was accepted by Plant Operations. The evaluation assumed bounding notching characteristics which were then evaluated for their impnt on the ability of systems to perform their specified functions. The notching characteristics assumed in the evaluation bound the notching observed during the October shutdow Therefore, the reactor water level instrumentation remains operabl o

.

i

~

6CIl0HS 00R!Nr. MCO #9:

Heasurement of instrument Dypass Valve Leakago The Reactor vessel reference legs were tested under TP92-47. This pressurized the reference ley and all instrument bypass valves to 20 psi. The test was performed on the 'A' and D' reference leg The 'B' side was completed first, and after stabilization, a leak rate of zero was determined af ter 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of tes On the 'A' side, pressure was applied between the root valve and the instrument bypass valve The total leskage observed was approximately 22 ml/ da However, leakage continued after all level transmitters had been valved ou The amount of leakage past the instrument bypass valves and how much was leaking past the normally open root valve could not ha determine No instrument bypass valves were replace Repair of External Leakaga The external leaks on the 'A' reference leg and 'B' reference leg were quentified prior to shutdown as approximately .14 ml/ day and 97 ml/ day, respectivel Additionally, seepage had been identified at mechanical joints but its too small to quanti fy. The identified leaks and seepage were tightened under a Maintenance Request and will be examined as part of Post-Work Testing af ter startu Condensing Chamber Steam Drain Line Insulation During a short outage in the spring of 1992, the insulation on the drain line (from condensing chamber to vessel) was removed from the 'B' sid It was originally postulated that this action could mitigate the water level anomolics noted during vessel depressurization. The insulation was removed to increase tho surface area available for condensation in order in increase the rate of steam flow into the condensing chamber. Although it was recognized that this would increase the reate of dilute gas flow into the drain line, it was expected the corresponding increase in condensate flow returning to the vessel would sweep / scrub a disproportionately larger amount of concentrated gas. This was expected to slow the rate of gas accumulation so as to delay the onset of or prevent gas binding of the chambe Unfortunately, the chamber temperature trends during operation and shutdown as well as the instrument performance during the shutdown indicata that little if any benefit was gained and the condensing chamber gas binding was not significantly retarde .

e m

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _

_ _ _ - - _ _ _ - _ _ _ _ _ - - -

. .

.. . . . . . . .-

.

s

"

However, during tho October 1992 shutdown, the pattern and volume of level errors on

'B' side instruments did not show any distinct improvement. Moreover, the insulation removal adds surface area for vapor condensation, potentially accelerating the inflow of dilute gas into the condensing chambers and-shortening the period for gas accumulation and concentr.ation in the chamber. Chamber temperature trends during the last operating period tend to support this possibilit Previous operation with the insulation on 'B' side did not result in . -

more adverse level anomalies than observed in the October 1992 shutdown. Thus, the reinstallation of this insulation during MCO #9 was implemente EYMll)Rl0fLOLJiOOT CAUSF ANALYSIS:

The instrumentation response was consistent with the non-condensible gas theory developed by Saul tevy, Inc. Additionally, ultrasonic detectors installed on the reference leg piping on the back of the 'B' ECCS Rack detected gases traveling u the piping during the spiking. Detectors on the 'A' side did not detect the gas

-

because the upper detector was not functionin Therefore, it is now believed that the cause of the spiking is due to non-iondensible gases coming out of solution during depressurization. As the gas forms a bubble and travels up the vertical sections of piping, the weight of the water in the reference leg is reducea and the transmitter indicates a higher level than actual. When the gas bubble reaches a horizontal section of piping, it has negligible effect on the level transmitter output and indicated level returns to actual level. The characteristic " square wave" shape is a function of the geometry of the reference leg piping. At low reactor pressure, gas is being released from multiple collection points at slightly different times. Therefore, the response takes on a ragged mountain peak shap Three factors are necessary for the spiking to occur. These are: (1) a buildup of non condensible gases in the condensing chanber; (2) horizontal sections in the reference leg piping; and (3) leakage near the transmitters which cause the gas saturated water to migrate down the reference leg. Elimination of any one of.these three factors should prevent the spiking during normal reactor depressurizatio IMPACT ON GENERIC LETTER 92-04 RESEQEE:

GL 92-04 requested licensees to determine the impact of level errors caused by non-condensible gas on automatic safety system response, Operator short-and long term actions, and Operator actions described-in Emergency Operating Procedures. Boston Edison performed a Plant Specific Safety Assessment in addition to the Generic

- Safety Assessment performed by the BWR Owner's Group. This safety assessment was reviewed by the DRC and approved by the Plant Department Manager. This assessment concluded that potential errors would have no affect on limiting FSAR transient and accident analyses, and in the event Operators are unable to determine water-leve E0Ps provide clear direction for Operator actions. The notching caused by non-condensible gas does not affect operability as. described previously, nor does_ it affect the conclusions of the safety assessment contained in Boston Edison's response to Gl92-0 , .

. - - . . . . . .-

,

.

The testing and analysis program being performed by the UWR0r. is predicated upon the existence of non condensible gas in reference legs. The notchly observed durin ,

this shutdown confirms non-condensible gas buildup. The observed notching was g consistent with our past exportence and does not invalidate our response to Generic letter 92 04. Therefore, the proper course of action is to pursue the BWROG testing and analysis program so the causes and ef fects of gas buildup can ho determined and fixed, if necessar OPIMBilllY_1YALW1110tl ron 2/3_ core covtRActfrmilssivt:

-

Since 2/3 core coverage interlock is the only level device that must function af ter reactor pressure has dropped to the point where dogassing occurs, the operability of this function has been revisited in light of data obtained during the recent shutdown and Industry information. Experience from the recent shutdown and provious shutdowns provides empirical evidence that notching of 29 inches or more can occu The root causa of level notching is now believed to be primarily due to non-condensibles. Although previous level responses during slow depressurizations reasonably compare with theoretically predicted level system res)onses, no validated

-

industry analyses, plant specific analyses have been completed t1at can prodlct responses during rapid depressurizations. Theoretical performance must be benchmarked against real test data. Such testing is actively being sursued by the

.

industry, lherefore, the magnitudo of any error can not accurately ao determined by analysis, it should be noted that over 20 linear feet of reference log volume, including both horizontal and vertical sections, must be lost during a rapid depressurization to create 14 inches of continuous level error. This level error is aircady considered in establishing the set point for the 2/3 core coverage lovel switc The potential for level errors should be considered lessened because the prosence of-non-condensibles rolles on transport into the reference legs dua to leakag External leakage is believed to be reduced from the leakage conditions that existed

_

during past operating portods. This eliminates one of the conditions' necessary for rapid depressurization level error as discussed in NRC Generic Letter 92-04, less gas in the reference legs should decrease nagnitude of the error due tn rapid

depressurizati6 The 2/3 core coverago condition is only expected for large recirculation pipo breaks. These breaks lead to level recovery to at least 2/3 core height'since this is the level of the refloodable vessel volume. The 2/3 core coverage permissivo prevents operators from inadvertently diverting LPCI flow to containment cooling if reactor levol is less than 2/3 core holght. Also, if LpCI flow has been diverted to containment l cooling and reactor level drops to 2/3 core height, thn containmnnt cooling valvos automatically close to redirect all tpCI flow to the reacto E0Ps direct operators to maintain level above top of active fuel (TAF). 2/3 core coverage is 4 feet below TAT. Level errors of up to approximately 4 feet would not prevent operators from maintaining 2/3 core coverage. This level is adequate for i coro coolirg.

!

.

-4-

!

=- -

- - -

r

..

.

Tech Spec lable 3.2.8 indicates that the 2/3 core coverane interlock prevents inadvertent actuation of containment spray during accident conditions. The trip device is set at the trip setting specified in the Labic. . Water level errors do not affect the sensed level at which the Tech Spec switches actuate and Table 3,2.B is therefore satisfied. However, Tech Spec Hases 3.2 indicates that the objective of the specification is to prescribe the trip setting required to assure adequate performanc As shown below, under rapid depressurizations, the system performance is adequate to a:sure the core cooling function. This condition, therefore, represents a potential nonconformance with the FSAR design- criteria for the system '

and not a noncompliance with Tech Specs. Water level inputs to the 2/3 core coverage interlock function are expected to the extent practical and feasible, to be true measures of operational conditions and to respond correctly to the sensed condition over the expected range of magnitudes and rates of change. Level errors represent a departure from fully conforming with these criteri This departure does not prevent the 2/3-core coverage interlock from detecting abnormal conditions with sufficient timeliness, precision, and reliability to assure that its specified safety functions can be performe Flow flow, diverson to containment spray represents roughly- 257, of a single RHR pump's lhe remaining LPCI flow to the reactor would still be well in excess of jet pump leakage (800 gpm per FSAR 3.3.6.5.2) and 2/3 core coverage would be maintained during spray diversion, it is also important to note that for large recirculation line breaks, at least one core spray loop will be available assuming a worst single failur Accounting for 800 gpm jet pump leakage, one core spray pump can maintain 2/3 core coverage and the effects of water level errors on the core cooling function are further diminishe A3pendix K LOCA Analyses for PNPS indicate that massive core uncovery (e.g., more t1an 10 feet below TAF) for approxim9ely 60 seconds can occur without exceeding 2200*F peak clad temperatures. The LOCA analysis is done assuming the reactor has just been shutdown due to the accident when fuel sensible heat and decay heat are still-high and coolant temperature is high. Worst peak clad temperatures are for the recirculation suction line break with LPCI Injection valve failur Conservative analysis indicates this peak clad temperature to be 1821*F which provides margin to 2200*F. Operator actions to divert LPCI flow would not occur until the core had been reflooded with comparatively cool water and fuel sensible and decay heats had substantially decreased. Loss of injection at this time would lead to level reduction due to bolloffs and not due to depressurization blowdow Core- spray flow- and some LPCI flow will maintain 2/3 core height. Multiple failures must occur to cause a condition where no core makeup flow exists, tevel errors of-up to five feet under this unlikely scenario will not lead to PCTs exceeding those in current LOCA analyse .

e a

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

r

,

'

The FSAR acknowledges somo level orrors occur during design basis event In selecting reactor water level as an input parameter to initiating safety functions, it is understood that some inaccuracy is inherent in the design. Deviations between actual and sensed lovel are expected due to pressure and thermal density effects, reference leg flashing, coolant voiding, and-flow across instrument nozzles (see FSAR 7.8.5.2). In selecting level trip settings, FSAR 7.2.4 states that the " trip point selected is not the only value of the trip point which results in acceptable results relative to tho fuel or nuclear system process barrie Trip setting selection is based on operating experience and is constrained by the safety. design basis". As has been shown, trip actuations at lovels different than assumed in accident analysis due to level errors do net load to unacceptable result The potential for loval error following a rapid depressurization is an industry-wide issue. Although the exact nature of the offects rapid depressurizations have.on the 2/3 core coverage interlock has not been quantified by the industry (BWROG) or Boston Edison, the available design margins, empirical plant data and reductions in reference leg leaks provide reasonable assurance that the P/3 core coverage interlock will remain operable, performed By:_ mA ll)I / T2 _

Reviewed By:_ .M Fo _29.dd_Co<C?er IEc;n rwi M C D Recommends Approval: >$ W t Rt/ P Nuc7/ar krig ing Managjin Recommends Approval: / _g/

l/_

_ _ _

e t go L Recommends Approval: . . . . .

GRC~Chairma /

ORC Meeting Number 9;L - too ._

Plant Manager Approval: d s m-6-